Support Vector Data Description

By C. Srivastava. 2011

1,952 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Continual Learning by Using Information of Each Class Holistically (2021, background): “Since PCL learns one class at a time, it is related to one-class learning (Sch¨ olkopf et al. 2001; Tax and Duin 2004; Hu et al. 2020; Perera, Nallapati, and Xiang 2019; Ruff et al. 2018).”
DOC: Deep Open Classification of Text Documents (2017, background): “DOC is markedly better than OpenMax and cbsSVM in macro-F 1 scores for both datasets in the 25%, 50%, and 75% settings.”
Recent Advances in Open Set Recognition: A Survey (2018, methodology): “Popular approaches for one-class classiﬁcation include one-class SVM [33] and support vector data description (SVDD) [35], [42], where one-class SVM separates the training samples from the origin of the feature space with a maximum margin, while SVDD encloses the training data with a hypersphere of ”
Deep Industrial Image Anomaly Detection: A Survey (2023, unknown): “And Sauter et al. [22] attempt to use the Xception network for classification and obtained results comparable to SVDD. FCDD [34] employs a fully convolutional neural network for OCC.”
Timeseries Anomaly Detection using Temporal Hierarchical One-Class Network (2020, methodology): “Two well-known and closely related one-class classiﬁcation models are the one-class support vector machine (OC-SVM) [26], which uses a hyperplane to separate the normal data from anomalous data; and the support vector data description (SVDD) [29], which uses a hypersphere to enclose the normal data.”
Unsupervised Deep Anomaly Detection for Multi-Sensor Time-Series Signals (2021, background): “For instance, One-Class Support Vector Machine (OCSVM) [19] and Support Vector Data Description (SVDD) [32] seek to learn a discriminative hypersphere surrounding the normal samples and then classify new data as normal or abnormal.”
Learning Deep Features for One-Class Classification (2018, methodology): “Based on stored template, a suitable one-class classiﬁer, such as one-class SVM [33], SVDD [35] or k-nearest neighbor, can be trained on the templates.”
Toward Supervised Anomaly Detection (2014, methodology): “Support vector learning has also been extended to many non-standard settings such as one-class learning (Schölkopf et al., 2001) and support vector data description (Tax & Duin, 2004).”
Local Novelty Detection in Multi-class Recognition Problems (2015, methodology): “The most popular techniques for one-class classiﬁcation are one-class SVM (1–SVM) [33] and support vector data description (SVDD) [34].”
Anomaly detection: A survey (2009, unknown): “A variant of the basic technique [Tax and Duin 1999a, 1999b; Tax 2001] .nds the smallest hypersphere in the kernel space that contains all training instances, and then determines on which side of that hypersphere a test instance lies.”
DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detection (2023, methodology): “[32], CL-MPPCA [67]; the density-estimation models: LOF [10], MPPCACD [76], DAGMM [92]; the clustering-based methods: DeepSVDD [60], THOC [64], ITAD [65]; the classic methods: OCSVM [68], OCSVM-based subsequence clustering (OCSVM*), IForest [46], IForest-based subsequence clustering (IForest*), Grad”
A Least-squares Approach to Direct Importance Estimation (2009, methodology): “Outlier Detection: The outlier detection task addressed here is to identify irregular samples in a validation data set based on a model data set that only contains regular samples (Scḧolkopf et al., 2001; Tax and Duin, 2004; Hodge and Austin, 2004; Hido et al., 2008).”
Learning Neural Random Fields with Inclusive Auxiliary Generators (2018, methodology): “From the results shown in Table 6, it can be seen that inclusive-NRF performs much better state-of-the-art method (DSVDD) which is specifically designed for anomaly detection.”
Statistical Analysis of Nearest Neighbor Methods for Anomaly Detection (2019, methodology): “Many empirical methods have been developed in the unsupervised setup, which can be roughly classified into four categories: density based methods such as the Robust KDE (RKDE) [4], Local Outlier Factor (LOF) [5], and mixture models (EGMM); distance based methods such as kNN [6] and Angle-based Outli”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, methodology): “Classic approaches to AD include PCA [1]–[5], the OC-SVM [6], SVDD [7], nearest neighbor algorithms [8]–[10], and KDE [11], [12].”
One-class classification: taxonomy of study and review of techniques (2013, unknown): “…Detection (Zhao et al ., 2005); Relevant Sentence Extraction (Kruengkrai & Jaruskulchai, 2003); Machine Vibration Analysis (Tax et al ., 1999); Machine Fault Detection (Ercil & Buke, 2002; Tax & Duin, 2004; Sarmiento et al ., 2005; Shin et al ., 2005); and Recommendation Tasks (Yasutoshi, 2006).”