Filters








38,037 Hits in 3.0 sec

Unsupervised Noisy Tracklet Person Re-identification [article]

Minxian Li, Xiatian Zhu, Shaogang Gong
2021 arXiv   pre-print
In this work, we present a novel selective tracklet learning (STL) approach that can train discriminative person re-id models from unlabelled tracklet data in an unsupervised manner.  ...  Importantly, our method is particularly more robust against arbitrary noisy data of raw tracklets therefore scalable to learning discriminative models from unconstrained tracking data.  ...  This data adaptive and selective matching capability is highly desired for dealing with noisy raw tracklets in unsupervised tracklet re-id learning.  ... 
arXiv:2101.06391v1 fatcat:ujxx5y4eincnleddi2ylq7wkqu

Robust Unsupervised Feature Selection on Networked Data

Jundong Li, Xia Hu, Liang Wu, Huan Liu
2016 Proceedings of the 2016 SIAM International Conference on Data Mining  
To address the above mentioned issues, in this paper, we propose a robust unsupervised feature selection framework NetFS for networked data, which embeds the latent representation learning into feature  ...  Feature selection has shown its effectiveness to prepare high-dimensional data for many data mining and machine learning tasks.  ...  Unsupervised Feature Selection for Networked Data Given: The feature set F, content matrix X and adjacency matrix A for all n instances.  ... 
doi:10.1137/1.9781611974348.44 dblp:conf/sdm/LiHWL16 fatcat:rnzy3xakkrasbn3a2tl73zd6ve

Feature weighting as a tool for unsupervised feature selection

Deepak Panday, Renato Cordeiro de Amorim, Peter Lane
2018 Information Processing Letters  
In this paper we introduce two unsupervised feature selection algorithms.  ...  Feature selection is a popular data pre-processing step.  ...  Section 3 introduces our new methods for unsupervised feature selection.  ... 
doi:10.1016/j.ipl.2017.09.005 fatcat:xjgygxzmvfgrlm6lwaeiuu3ajy

Unsupervised Feature Selection for Outlier Detection by Modelling Hierarchical Value-Feature Couplings

Guansong Pang, Longbing Cao, Ling Chen, Huan Liu
2016 2016 IEEE 16th International Conference on Data Mining (ICDM)  
This paper proposes a novel Coupled Unsupervised Feature Selection framework (CUFS for short) to filter out noisy or redundant features for subsequent outlier detection in categorical data.  ...  Proper feature selection for unsupervised outlier detection can improve detection performance but is very challenging due to complex feature interactions, the mixture of relevant features with noisy/redundant  ...  Little work has been designed to conduct feature selection for unsupervised outlier detection in categorical data.  ... 
doi:10.1109/icdm.2016.0052 dblp:conf/icdm/PangCCL16 fatcat:nxekvj6ijveadcsli2w22447oi

Differentiable Unsupervised Feature Selection based on a Gated Laplacian [article]

Ofir Lindenbaum, Uri Shaham, Jonathan Svirsky, Erez Peterfreund, Yuval Kluger
2020 arXiv   pre-print
In this paper, we present a method for unsupervised feature selection, and we demonstrate its use for the task of clustering.  ...  We propose a differentiable loss function that combines the Laplacian score, which favors low-frequency features, with a gating mechanism for feature selection.  ...  Acknowledgements The authors thank Stefan Steinerberger, Boaz Nadler and Ronen Basri for helpful discussions.  ... 
arXiv:2007.04728v3 fatcat:2gvrlo46jjbnbijd2tbioylive

Clustering-Based Band Selection Using Structural Similarity Index and Entropy for Hyperspectral Image Classification

Arsalan Ghorbanian, Yasser Maghsoudi, Ali Mohammadzadeh
2020 Traitement du signal  
In this paper, an unsupervised Feature Selection (FS) algorithm was proposed for hyperspectral image classification.  ...  Despite the unique capabilities of hyperspectral images for classification tasks, handling the high dimension of these data is challenging.  ...  CONCLUSION This study presents an unsupervised FS method for hyperspectral band selection. Entropy and the SSIM index were employed to remove noisy bands and select informative bands, respectively.  ... 
doi:10.18280/ts.370510 fatcat:zxeqxbxjrjehxgqas7jumrjgke

Robust Spectral Learning for Unsupervised Feature Selection

Lei Shi, Liang Du, Yi-Dong Shen
2014 2014 IEEE International Conference on Data Mining  
In this paper, we consider the problem of unsupervised feature selection.  ...  In this paper, we propose a Robust Spectral learning framework for unsupervised Feature Selection (RSFS), which jointly improves the robustness of graph embedding and sparse spectral regression.  ...  ACKNOWLEDGMENTS We would like to thank the anonymous reviewers for their helpful comments and suggestions.  ... 
doi:10.1109/icdm.2014.58 dblp:conf/icdm/ShiDS14 fatcat:6nr5yqyranemjpkwjqlsy2ihoe

MFSPFA: An Enhanced Filter based Feature Selection Algorithm

V. ArulKumar, L. Arockiam
2012 International Journal of Computer Applications  
Feature Selection is the process of selecting the momentous feature subset from the original ones. This technique is frequently used as a preprocessing technique in data mining.  ...  In this study, a new feature selection algorithm is proposed and is called Modified Fisher Score Principal Feature Analysis (MFSPFA).  ...  INTRODUCTION Feature Selection is the process of selecting the subset of relevant features by removing redundant, irrelevant and noisy data from the original dataset.  ... 
doi:10.5120/8096-1682 fatcat:3bet63ziorfnjgeyx6ete7rap4

Noise-Resistant Unsupervised Feature Selection via Multi-perspective Correlations

Hao Huang, Shinjae Yoo, Dantong Yu, Hong Qin
2014 2014 IEEE International Conference on Data Mining  
Unsupervised feature selection is an important issue for high dimensional dataset analysis. However popular methods are susceptible to noisy instances (observations) or noisy features.  ...  Our proposed approach, called Noise-Resistant Unsupervised Feature Selection (NRFS), is based on multi-perspective correlation that reflects the importance of feature with respect to noise-resistant representative  ...  Abstract-Unsupervised feature selection is an important issue for high dimensional dataset analysis. However popular methods are susceptible to noisy instances (observations) or noisy features.  ... 
doi:10.1109/icdm.2014.88 dblp:conf/icdm/HuangYYQ14a fatcat:hkx7wtr26ja3bbmve6quqheu7q

Unsupervised Feature Selection with Adaptive Structure Learning

Liang Du, Yi-Dong Shen
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
Traditional unsupervised methods select the features which can faithfully preserve the intrinsic structures of data, where the intrinsic structures are estimated using all the input features of data.  ...  Experimental results on many benchmark data sets demonstrate that the proposed method outperforms many state of the art unsupervised feature selection methods.  ...  Mingyu Fan for their helpful suggestions to improve this paper.  ... 
doi:10.1145/2783258.2783345 dblp:conf/kdd/DuS15 fatcat:lxg6miunrncgzmp4by352dklvy

Joint Noise-Tolerant Learning and Meta Camera Shift Adaptation for Unsupervised Person Re-Identification

Fengxiang Yang, Zhun Zhong, Zhiming Luo, Yuanzheng Cai, Yaojin Lin, Shaozi Li, Nicu Sebe
2021 Zenodo  
This paper considers the problem of unsupervised person re-identification (re-ID), which aims to learn discriminative models with unlabeled data.  ...  Although this kind of approach has shown promising accuracy, it is hampered by 1) noisy labels produced by clustering and 2) feature variations caused by camera shift.  ...  For the line of unsupervised learning, Hsu et al. [13] attempt to solve the few-shot problem with unlabeled meta-train data and labeled meta-test data.  ... 
doi:10.5281/zenodo.5014558 fatcat:hm4mo4jpandvfk2jfeq2sh26b4

Spectral ranking and unsupervised feature selection for point, collective, and contextual anomaly detection

Haofan Zhang, Ke Nian, Thomas F. Coleman, Yuying Li
2018 International Journal of Data Science and Analytics  
Accordingly, we propose an unsupervised backward elimination feature selection algorithm BAHSIC-AD, utilizing Hilbert-Schmidt Independence Critirion (HSIC) in identifying the data instances present as  ...  anomalies in the subset of features that have strong dependence with each other.  ...  Feature Selection for Anomaly Detection Extensive research has been conducted on the subject of supervised feature selection [22] [41] [31] , and many attempts have been made for the unsupervised clustering  ... 
doi:10.1007/s41060-018-0161-7 fatcat:xzbxwr3ujfckbnbi2bza2a4mqq

A Survey on Feature Selection Algorithms

Amit Kumar
2015 International Journal on Recent and Innovation Trends in Computing and Communication  
The paper surveys historic developments reported in feature selection with supervised and unsupervised methods.  ...  Due to its applications in several areas including data mining, soft computing and big data analysis, feature selection has got a reasonable importance.  ...  Feature selection and dimensionality reduction possess mostly a common goal which is to reduce the number of harmful, irrelevant and noisy features in a dataset for smooth and fast data processing purposes  ... 
doi:10.17762/ijritcc2321-8169.150431 fatcat:tme4tyl4ijaunbzraka3tey654

Local Selection of Features for Image Search and Annotation

Jichao Sun
2014 Proceedings of the ACM International Conference on Multimedia - MM '14  
We propose several techniques for the local selection of features for image databases.  ...  By checking the local neighborhood of each image, our methods determine the feature importance with respect to the image and select different feature sets for individual images.  ...  Laplacian Score (LS ), an unsupervised method for feature selection of generic data, ranks individual features according to their locality-preserving abilities [4] .  ... 
doi:10.1145/2647868.2654863 dblp:conf/mm/Sun14 fatcat:qy4uqwwknzaftdpk4lswe3a4pa

Unsupervised Feature Selection with Adaptive Structure Learning [article]

Liang Du, Yi-Dong Shen
2015 arXiv   pre-print
Traditional unsupervised methods select the features which can faithfully preserve the intrinsic structures of data, where the intrinsic structures are estimated using all the input features of data.  ...  Experimental results on many benchmark data sets demonstrate that the proposed method outperforms many state of the art unsupervised feature selection methods.  ...  Without class label, unsupervised feature selection chooses features that can effectively reveal or maintain the underlying structure of data.  ... 
arXiv:1504.00736v1 fatcat:w6int3ap6zd5plmubmdz4bihuu
« Previous Showing results 1 — 15 out of 38,037 results