The Experts below are selected from a list of 41235 Experts worldwide ranked by ideXlab platform
Yaohang Li - One of the best experts on this subject based on the ideXlab platform.
-
IJCAI - Single-Pass PCA of Large High-Dimensional Data.
Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, 2017Co-Authors: Wenjian Yu, Yu Gu, Jian Li, Yaohang LiAbstract:Principal component analysis (PCA) is a fundamental dimension reduction tool in statistics and machine learning. For large and High-Dimensional Data, computing the PCA (i.e., the top singular vectors of the Data matrix) becomes a challenging task. In this work, a single-pass randomized algorithm is proposed to compute PCA with only one pass over the Data. It is suitable for processing extremely large and High-Dimensional Data stored in slow memory (hard disk) or the Data generated in a streaming fashion. Experiments with synthetic and real Data validate the algorithm's accuracy, which has orders of magnitude smaller error than an existing single-pass algorithm. For a set of High-Dimensional Data stored as a 150 GB file, the algorithm is able to compute the first 50 principal components in just 24 minutes on a typical 24-core computer, with less than 1 GB memory cost.
-
Single-Pass PCA of Large High-Dimensional Data
arXiv: Data Structures and Algorithms, 2017Co-Authors: Wenjian Yu, Yu Gu, Jian Li, Yaohang LiAbstract:Principal component analysis (PCA) is a fundamental dimension reduction tool in statistics and machine learning. For large and High-Dimensional Data, computing the PCA (i.e., the singular vectors corresponding to a number of dominant singular values of the Data matrix) becomes a challenging task. In this work, a single-pass randomized algorithm is proposed to compute PCA with only one pass over the Data. It is suitable for processing extremely large and High-Dimensional Data stored in slow memory (hard disk) or the Data generated in a streaming fashion. Experiments with synthetic and real Data validate the algorithm's accuracy, which has orders of magnitude smaller error than an existing single-pass algorithm. For a set of High-Dimensional Data stored as a 150 GB file, the proposed algorithm is able to compute the first 50 principal components in just 24 minutes on a typical 24-core computer, with less than 1 GB memory cost.
Nenad Tomasev - One of the best experts on this subject based on the ideXlab platform.
-
The Role Of Hubness in High-Dimensional Data Analysis
Informatica (lithuanian Academy of Sciences), 2020Co-Authors: Nenad TomasevAbstract:This article presents a summary of the doctoral dissertation of the author, which addresses the task of machine learning under hubness in intrinsically High-Dimensional Data.
-
The Role of Hubness in Clustering High-Dimensional Data
IEEE Transactions on Knowledge and Data Engineering, 2014Co-Authors: Nenad Tomasev, Dunja Mladenic, Miloš Radovanović, Mirjana IvanovićAbstract:High-Dimensional Data arise naturally in many domains, and have regularly presented a great challenge for traditional Data mining techniques, both in terms of effectiveness and efficiency. Clustering becomes difficult due to the increasing sparsity of such Data, as well as the increasing difficulty in distinguishing distances between Data points. In this paper, we take a novel perspective on the problem of clustering High-Dimensional Data. Instead of attempting to avoid the curse of Dimensionality by observing a lower Dimensional feature subspace, we embrace Dimensionality by taking advantage of inherently High-Dimensional phenomena. More specifically, we show that hubness, i.e., the tendency of High-Dimensional Data to contain points (hubs) that frequently occur in k-nearest-neighbor lists of other points, can be successfully exploited in clustering. We validate our hypothesis by demonstrating that hubness is a good measure of point centrality within a High-Dimensional Data cluster, and by proposing several hubness-based clustering algorithms, showing that major hubs can be used effectively as cluster prototypes or as guides during the search for centroid-based cluster configurations. Experimental results demonstrate good performance of our algorithms in multiple settings, particularly in the presence of large quantities of noise. The proposed methods are tailored mostly for detecting approximately hyperspherical clusters and need to be extended to properly handle clusters of arbitrary shapes.
-
the role of hubness in clustering High Dimensional Data
Knowledge Discovery and Data Mining, 2011Co-Authors: Nenad Tomasev, Milos Radovanovic, Dunja Mladenic, Mirjana IvanovicAbstract:High-Dimensional Data arise naturally in many domains, and have regularly presented a great challenge for traditional Data-mining techniques, both in terms of effectiveness and efficiency. Clustering becomes difficult due to the increasing sparsity of such Data, as well as the increasing difficulty in distinguishing distances between Data points. In this paper we take a novel perspective on the problem of clustering High-Dimensional Data. Instead of attempting to avoid the curse of Dimensionality by observing a lower-Dimensional feature subspace, we embrace Dimensionality by taking advantage of some inherently High-Dimensional phenomena. More specifically, we show that hubness, i.e., the tendency of High-Dimensional Data to contain points (hubs) that frequently occur in k-nearest neighbor lists of other points, can be successfully exploited in clustering. We validate our hypothesis by proposing several hubness-based clustering algorithms and testing them on High-Dimensional Data. Experimental results demonstrate good performance of our algorithms in multiple settings, particularly in the presence of large quantities of noise.
-
PAKDD (1) - The role of hubness in clustering High-Dimensional Data
Advances in Knowledge Discovery and Data Mining, 2011Co-Authors: Nenad Tomasev, Dunja Mladenic, Miloš Radovanović, Mirjana IvanovićAbstract:High-Dimensional Data arise naturally in many domains, and have regularly presented a great challenge for traditional Data-mining techniques, both in terms of effectiveness and efficiency. Clustering becomes difficult due to the increasing sparsity of such Data, as well as the increasing difficulty in distinguishing distances between Data points. In this paper we take a novel perspective on the problem of clustering High-Dimensional Data. Instead of attempting to avoid the curse of Dimensionality by observing a lower-Dimensional feature subspace, we embrace Dimensionality by taking advantage of some inherently High-Dimensional phenomena. More specifically, we show that hubness, i.e., the tendency of High-Dimensional Data to contain points (hubs) that frequently occur in k-nearest neighbor lists of other points, can be successfully exploited in clustering. We validate our hypothesis by proposing several hubness-based clustering algorithms and testing them on High-Dimensional Data. Experimental results demonstrate good performance of our algorithms in multiple settings, particularly in the presence of large quantities of noise.
Beizhan Wang - One of the best experts on this subject based on the ideXlab platform.
-
The Key Data Mining Models for High Dimensional Data
Proceedings of the 2012 International Conference on Communication Electronics and Automation Engineering, 2020Co-Authors: Xiang Deng, Beizhan Wang, Minkui ChenAbstract:With the rapid development of computational biology and e-commerce applications, High-Dimensional Data becomes more and more powerful. Thus, it is an urgent problem of great importance when mining High-Dimensional Data. However, there are some challenges for mining Data of High dimensions, the first one is the curse of Dimensionality and the second one is the meaningfulness of the similarity measure in the High dimension space. In this paper, we present several state-of-art techniques for constructing three Data mining models with analyzing High-Dimensional Data, these models include frequent pattern mining, clustering, and classification. And we discuss how these methods deal with the challenges of High Dimensionality.
-
An Initialization Method for Clustering High-Dimensional Data
2009 First International Workshop on Database Technology and Applications, 2009Co-Authors: Luying Chen, Lifei Chen, Qingshan Jiang, Beizhan WangAbstract:In iterative refinement clustering algorithms, such as the various types of K-Means algorithms, the clustering results are very sensitive to the initial cluster centers. Conventional initialization methods tend to loss effectiveness due to the so-called "curse of Dimensionality" when clustering High-Dimensional Data. In this paper, a local density based method is proposed to search for initial cluster centers on High-Dimensional Data. We define the probability density of a point as the amount of its Highly similar neighborhoods with weight coefficient. Points with High density neighborhoods and low similarity are chosen as the initial cluster centers. Experimental results on real world Datasets show the effectiveness of the proposed method.
-
DBTA - An Initialization Method for Clustering High-Dimensional Data
2009 First International Workshop on Database Technology and Applications, 2009Co-Authors: Luying Chen, Lifei Chen, Qingshan Jiang, Beizhan WangAbstract:In iterative refinement clustering algorithms, such as the various types of K-Means algorithms, the clustering results are very sensitive to the initial cluster centers. Conventional initialization methods tend to loss effectiveness due to the so-called "curse of Dimensionality" when clustering High-Dimensional Data. In this paper, a local density based method is proposed to search for initial cluster centers on High-Dimensional Data. We define the probability density of a point as the amount of its Highly similar neighborhoods with weight coefficient. Points with High density neighborhoods and low similarity are chosen as the initial cluster centers. Experimental results on real world Datasets show the effectiveness of the proposed method
Wenjian Yu - One of the best experts on this subject based on the ideXlab platform.
-
IJCAI - Single-Pass PCA of Large High-Dimensional Data.
Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, 2017Co-Authors: Wenjian Yu, Yu Gu, Jian Li, Yaohang LiAbstract:Principal component analysis (PCA) is a fundamental dimension reduction tool in statistics and machine learning. For large and High-Dimensional Data, computing the PCA (i.e., the top singular vectors of the Data matrix) becomes a challenging task. In this work, a single-pass randomized algorithm is proposed to compute PCA with only one pass over the Data. It is suitable for processing extremely large and High-Dimensional Data stored in slow memory (hard disk) or the Data generated in a streaming fashion. Experiments with synthetic and real Data validate the algorithm's accuracy, which has orders of magnitude smaller error than an existing single-pass algorithm. For a set of High-Dimensional Data stored as a 150 GB file, the algorithm is able to compute the first 50 principal components in just 24 minutes on a typical 24-core computer, with less than 1 GB memory cost.
-
Single-Pass PCA of Large High-Dimensional Data
arXiv: Data Structures and Algorithms, 2017Co-Authors: Wenjian Yu, Yu Gu, Jian Li, Yaohang LiAbstract:Principal component analysis (PCA) is a fundamental dimension reduction tool in statistics and machine learning. For large and High-Dimensional Data, computing the PCA (i.e., the singular vectors corresponding to a number of dominant singular values of the Data matrix) becomes a challenging task. In this work, a single-pass randomized algorithm is proposed to compute PCA with only one pass over the Data. It is suitable for processing extremely large and High-Dimensional Data stored in slow memory (hard disk) or the Data generated in a streaming fashion. Experiments with synthetic and real Data validate the algorithm's accuracy, which has orders of magnitude smaller error than an existing single-pass algorithm. For a set of High-Dimensional Data stored as a 150 GB file, the proposed algorithm is able to compute the first 50 principal components in just 24 minutes on a typical 24-core computer, with less than 1 GB memory cost.
Magnus Jansson - One of the best experts on this subject based on the ideXlab platform.
-
ACSSC - Model selection for High-Dimensional Data
2016 50th Asilomar Conference on Signals Systems and Computers, 2016Co-Authors: Arash Owrang, Magnus JanssonAbstract:We investigate the task of model selection for High-Dimensional Data. For this purpose, we propose an extension to the Bayesian information criterion. Our information criterion is asymptotically consistent either as the number of measurements tends to infinity or as the variance of noise decreases to zero. The numerical results provided support our claim. Additionally, we Highlight the link between model selection for High-Dimensional Data and the choice of hyper-parameter in l 1 -constrained estimators, specifically the LASSO.
-
Model selection for High-Dimensional Data
2016 50th Asilomar Conference on Signals Systems and Computers, 2016Co-Authors: Arash Owrang, Magnus JanssonAbstract:We investigate the task of model selection for High-Dimensional Data. For this purpose, we propose an extension to the Bayesian information criterion. Our information criterion is asymptotically consistent either as the number of measurements tends to infinity or as the variance of noise decreases to zero. The numerical results provided support our claim. Additionally, we Highlight the link between model selection for High-Dimensional Data and the choice of hyper-parameter in ℓ1-constrained estimators, specifically the LASSO.