Search For:

Displaying 1-25 out of 25 total
A Comparative Study of Linear and Nonlinear Feature Extraction Methods
Found in: Data Mining, IEEE International Conference on
By Cheong Hee Park, Haesun Park, Panos Pardalos
Issue Date:November 2004
pp. 495-498
This paper presents theoretical relationships among several generalized LDA algorithms and proposes computationally efficient approaches for them utilizing the relationships. Generalized LDA algorithms are extended nonlinearly by kernel methods resulting i...
 
UTOPIAN: User-Driven Topic Modeling Based on Interactive Nonnegative Matrix Factorization
Found in: IEEE Transactions on Visualization and Computer Graphics
By Jaegul Choo,Changhyun Lee,Chandan K. Reddy,Haesun Park
Issue Date:December 2013
pp. 1992-2001
Topic modeling has been widely used for analyzing text document collections. Recently, there have been significant advancements in various topic modeling techniques, particularly in the form of probabilistic graphical modeling. State-of-the-art techniques ...
 
A new optimization criterion for generalized discriminant analysis on undersampled problems
Found in: Data Mining, IEEE International Conference on
By Jieping Ye, Ravi Janardan, Cheong Hee Park, Haesun Park
Issue Date:November 2003
pp. 419
A new optimization criterion for discriminant analysis is presented. The new criterion extends the optimization criteria of the classical linear discriminant analysis (LDA) by introducing the pseudo-inverse when the scatter matrices are singular. It is app...
 
Efficient Nonlinear Dimension Reduction for Clustered Data Using Kernel Functions
Found in: Data Mining, IEEE International Conference on
By Cheong Hee Park, Haesun Park
Issue Date:November 2003
pp. 243
In this paper, we propose a nonlinear feature extraction method which is based on centroids and kernel functions. The dimension reducing nonlinear transformation is obtained by implicitly mapping the input data into a feature space using a kernel function,...
 
Interactive Data Analysis Tool by Augmenting MATLAB with Semantic Objects
Found in: 2013 IEEE 13th International Conference on Data Mining Workshops (ICDMW)
By Changhyun Lee,Jaegul Choo,Duen Horng Polo Chau,Haesun Park
Issue Date:December 2013
pp. 1093-1096
The traditional data analysis tools support strong computational capabilities and numerous standard visualization techniques. However, they provide little visual interactions due to the fact that the tools maintain a wide applicability to diverse data doma...
 
Regularization Paths for Sparse Nonnegative Least Squares Problems with Applications to Life Cycle Assessment Tree Discovery
Found in: 2013 IEEE International Conference on Data Mining (ICDM)
By Jingu Kim,Naren Ramakrishnan,Manish Marwah,Amip Shah,Haesun Park
Issue Date:December 2013
pp. 360-369
The nonnegative least squares problems are useful in applications where the physical nature of problem domain permits only additive linear combinations. We discuss the l1-regularized nonnegative least squares (L1-NLS) problem, where l1-regularization is us...
 
Combining Computational Analyses and Interactive Visualization for Document Exploration and Sensemaking in Jigsaw
Found in: IEEE Transactions on Visualization and Computer Graphics
By C. Gorg, Zhicheng Liu, Jaeyeon Kihm, Jaegul Choo, Haesun Park,J. Stasko
Issue Date:October 2013
pp. 1646-1663
Investigators across many disciplines and organizations must sift through large collections of text documents to understand and piece together information. Whether they are fighting crime, curing diseases, deciding what car to buy, or researching a new fie...
 
Bounded Matrix Low Rank Approximation
Found in: 2012 IEEE 12th International Conference on Data Mining (ICDM)
By Ramakrishnan Kannan,Mariya Ishteva,Haesun Park
Issue Date:December 2012
pp. 319-328
Matrix lower rank approximations such as non-negative matrix factorization (NMF) have been successfully used to solve many data mining tasks. In this paper, we propose a new matrix lower rank approximation called Bounded Matrix Low Rank Approximation (BMA)...
 
Toward Faster Nonnegative Matrix Factorization: A New Algorithm and Comparisons
Found in: Data Mining, IEEE International Conference on
By Jingu Kim, Haesun Park
Issue Date:December 2008
pp. 353-362
Nonnegative Matrix Factorization (NMF) is a dimension reduction method that has been widely used for various tasks including text mining, pattern analysis, clustering, and cancer class discovery. The mathematical formulation for NMF appears as a non-convex...
 
A Comparison of Unsupervised Dimension Reduction Algorithms for Classification
Found in: Bioinformatics and Biomedicine, IEEE International Conference on
By Jaegul Choo, Hyunsoo Kim, Haesun Park, Hongyuan Zha
Issue Date:November 2007
pp. 71-77
Distance preserving dimension reduction (DPDR) using the singular value decomposition has recently been intro- duced. In this paper, for disease diagnosis using gene or protein expression data, we present empirical comparison results between DPDR and other...
 
Feature Reduction via Generalized Uncorrelated Linear Discriminant Analysis
Found in: IEEE Transactions on Knowledge and Data Engineering
By Jieping Ye, Ravi Janardan, Qi Li, Haesun Park
Issue Date:October 2006
pp. 1312-1322
High-dimensional data appear in many applications of data mining, machine learning, and bioinformatics. Feature reduction is commonly applied as a preprocessing step to overcome the curse of dimensionality. Uncorrelated Linear Discriminant Analysis (ULDA) ...
 
Adaptive Nonlinear Discriminant Analysis by Regularized Minimum Squared Errors
Found in: IEEE Transactions on Knowledge and Data Engineering
By Hyunsoo Kim, Barry L. Drake, Haesun Park
Issue Date:May 2006
pp. 603-612
Kernelized nonlinear extensions of Fisher's discriminant analysis, discriminant analysis based on generalized singular value decomposition (LDA/GSVD), and discriminant analysis based on the minimum squared error formulation (MSE) have recently been widely ...
 
IDR/QR: An Incremental Dimension Reduction Algorithm via QR Decomposition
Found in: IEEE Transactions on Knowledge and Data Engineering
By Jieping Ye, Qi Li, Hui Xiong, Haesun Park, Ravi Janardan, Vipin Kumar
Issue Date:September 2005
pp. 1208-1222
Dimension reduction is a critical data preprocessing step for many database and data mining applications, such as efficient storage and retrieval of high-dimensional data. In the literature, a well-known dimension reduction algorithm is Linear Discriminant...
 
Generalizing Discriminant Analysis Using the Generalized Singular Value Decomposition
Found in: IEEE Transactions on Pattern Analysis and Machine Intelligence
By Peg Howland, Haesun Park
Issue Date:August 2004
pp. 995-1006
<p><b>Abstract</b>—Discriminant analysis has been used for decades to extract features that preserve class separability. It is commonly defined as an optimization problem involving covariance matrices that represent the scatter within and...
 
Imputation of Missing Values in DNA Microarray Gene Expression Data
Found in: Computational Systems Bioinformatics Conference, International IEEE Computer Society
By Hyunsoo Kim, Gene H. Golub, Haesun Park
Issue Date:August 2004
pp. 572-573
Most multivariate statistical methods for gene expression data require a complete matrix of gene array values. In this paper, a imputation method based on least squares formulation is proposed to estimate missing values. It exploits local similarity struct...
   
Incremental and Decremental Least Squares Support Vector Machine and Its Application to Drug Design
Found in: Computational Systems Bioinformatics Conference, International IEEE Computer Society
By Hyunsoo Kim, Haesun Park
Issue Date:August 2004
pp. 656-657
The least squares support vector machine (LS-SVM) has shown to exhibit excellent classification performance in many applications. In this paper, we propose an incremental and decremental LS-SVM based on updating and downdating the QR decomposition. It can ...
   
UTOPIAN: User-Driven Topic Modeling Based on Interactive Nonnegative Matrix Factorization
Found in: IEEE Transactions on Visualization and Computer Graphics
By Jaegul Choo, Changhyun Lee,Chandan K. Reddy,Haesun Park
Issue Date:December 2013
pp. 1992-2001
Topic modeling has been widely used for analyzing text document collections. Recently, there have been significant advancements in various topic modeling techniques, particularly in the form of probabilistic graphical modeling. State-of-the-art techniques ...
 
Customizing Computational Methods for Visual Analytics with Big Data
Found in: IEEE Computer Graphics and Applications
By Jaegul Choo, Haesun Park
Issue Date:July 2013
pp. 22-28
The volume of available data has been growing exponentially, increasing data problem's complexity and obscurity. In response, visual analytics (VA) has gained attention, yet its solutions haven't scaled well for big data. Computational methods can improve ...
 
Structured total least norm method for Toeplitz problems
Found in: Acoustics, Speech, and Signal Processing, IEEE International Conference on
By Haesun Park, J. Ben Rosen, J. Glick
Issue Date:May 1995
pp. 1141-1144
The total least squares (TLS) method for solving an overdetermined system Ax/spl ap/b is a generalization of the least squares (LS) method, and it minimizes /spl par/[E|r]/spl par//sub F/ so that (b+r)/spl isin/Range(A+E), given A/spl isin/R/sup mxn/, with...
 
Fast interactive visualization for multivariate data exploration
Found in: CHI '13 Extended Abstracts on Human Factors in Computing Systems (CHI EA '13)
By Changhyun Lee, Duen Horng (Polo) Chau, Haesun Park, Jaegul Choo, Wei Zhuo
Issue Date:April 2013
pp. 1773-1778
We are investigating a fast layout method for visualizing and exploring relationships between multivariate data items. We improve on existing works that use the force-directed layout, which has high running time and cannot scale up for large-scale visual a...
     
Fast bregman divergence NMF using taylor expansion and coordinate descent
Found in: Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '12)
By Guy Lebanon, Haesun Park, Liangda Li
Issue Date:August 2012
pp. 307-315
Non-negative matrix factorization (NMF) provides a lower rank approximation of a matrix. Due to nonnegativity imposed on the factors, it gives a latent structure that is often more physically meaningful than other lower rank approximations such as singular...
     
Orthogonal nonnegative matrix t-factorizations for clustering
Found in: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '06)
By Chris Ding, Haesun Park, Tao Li, Wei Peng
Issue Date:August 2006
pp. 126-135
Currently, most research on nonnegative matrix factorization (NMF)focus on 2-factor $X=FG^T$ factorization. We provide a systematicanalysis of 3-factor $X=FSG^T$ NMF. While it unconstrained 3-factor NMF is equivalent to it unconstrained 2-factor NMF, itcon...
     
IDR/QR: an incremental dimension reduction algorithm via QR decomposition
Found in: Proceedings of the 2004 ACM SIGKDD international conference on Knowledge discovery and data mining (KDD '04)
By Haesun Park, Hui Xiong, Jieping Ye, Qi Li, Ravi Janardan, Vipin Kumar
Issue Date:August 2004
pp. 364-373
Dimension reduction is critical for many database and data mining applications, such as efficient storage and retrieval of high-dimensional data. In the literature, a well-known dimension reduction scheme is Linear Discriminant Analysis (LDA). The common a...
     
Feature extraction via generalized uncorrelated linear discriminant analysis
Found in: Twenty-first international conference on Machine learning (ICML '04)
By Haesun Park, Jieping Ye, Qi Li, Ravi Janardan
Issue Date:July 2004
pp. 182-182
Feature extraction is important in many applications, such as text and image retrieval, because of high dimensionality. Uncorrelated Linear Discriminant Analysis (ULDA) was recently proposed for feature extraction. The extracted features via ULDA were show...
     
An algorithm for the generalized singular value decomposition on massively parallel computers
Found in: Proceedings of the 5th international conference on Supercomputing (ICS '91)
By Haesun Park, L. Magnus Ewerbring
Issue Date:June 1991
pp. 136-145
The paper traces the development of large multi-transputer systems for high-performance scientific and engineering computing. After defining what we mean by 'supercomputing' in the context of this paper, the past and present state of transputer supercomput...
     
 1