A Classification Framework for Anomaly Detection

By Ingo Steinwart, Don R. Hush, Clint Scovel. 2005

359 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

Security analysis of online centroid anomaly detection (2010, methodology): “Hence several direct methods have been proposed for anomaly detection, e.g., oneclass SVM (Schölkopf et al., 2001), support vector data description (SVDD) (Tax and Duin, 1999a,b), and density level set estimation (Polonik, 1995; Tsybakov, 1997; Steinwart et al., 2005).”
How to Compare Different Loss Functions and Their Risks (2007, methodology): “To overcome this problem a framework was developed in [28] that translates the density level detection problem to a binary classification problem.”
Distribution Free Prediction Sets (2013, methodology): “3 Related work The conformal prediction method (Vovk et al. 2005; Shafer & Vovk 2008) is a general approach for constructing distribution free, sequential prediction sets using exchangeability, and is usually applied to sequential classification and regression problems (Vovk, Nouretdinov & Gammerman”
Surface street traffic estimation (2007, background): “Classifying unlabeled data is a well-known problem [21, 9] in the area of data mining and can be solved in general by clustering.”
Anomaly detection: A survey (2009, methodology): “Nearest Neighbor based Techniques Section 5 Subramaniam et al. [2006], Kejia Zhang and Li [2007], Idé et al.”
QP Algorithms with Guaranteed Accuracy and Run Time for Support Vector Machines (2006, methodology): “the DLD–SVM for the density level detection problem described in (Steinwart et al., 2005) which sets ui = {”
Semi-Supervised Novelty Detection (2010, methodology): “The standard approach has been to estimate a level set of the nominal density (Schölkopf et al., 2001; Steinwart et al., 2005; Scott and Nowak, 2006; Vert and Vert, 2006; El-Yaniv and Nisenson, 2007; Hero, 2007), and to declare test points outside the estimated level set to be novelties.”
Optimal rates for plug-in estimators of density level sets (2006, background): “, Steinwart et al. (2005)), they are still not very easy to use for practical purposes, compared to plug-in methods.”
Quantile Regression Forests (2006, background): “Other research has focused on detecting anomalies for unlabelled data (e.g. Markou and Singh, 2003; Steinwart et al., 2005).”
Ranking the Best Instances (2006, methodology): “Remark 3 (CONNECTION WITH REGRESSION LEVEL SET ESTIMATION) We mention that the estimation of the level sets of the regression function has been studied in the statistics literature (Cavalier, 1997) (see also Tsybakov, 1997 and Willett and Nowak, 2006) as well as in the learning literature, for insta”
Outlier detection by active learning (2006, background): “Recent work by Steinwart et al [15] gives a comprehensive treatment of the classification approach to anomaly detection, based on a theoretical framework of density-level detection.”
Outlier Detection (2008, background): “Tang et al. [55] defined the connectivity-based outlier factor that compares favorably with LOF at low density regions.”
A Unifying Review of Deep and Shallow Anomaly Detection (2020, methodology): “Some well-known unsupervised methods, such as KDE [12] or the OC-SVM [6], implicitly make this assumption that P follows a uniform distribution that can be interpreted as a default uninformative prior on the anomalous distribution [212].”
Unsupervised Learning Methods for Molecular Simulation Data (2021, unknown): “Although its main applications relate to data visualization (as in the case of visualizing the free energy surface), it is also part of the pipeline of other analysis methods such as kernel regression, 174 anomaly detection, 175 and clustering (see Section 5).”
Learning Sets with Separating Kernels (2012, methodology): “In this context, some algorithms inspired by Support Vector Machine (SVM) have been studied in [56, 64, 69].”
Deep Semi-Supervised Anomaly Detection (2019, background): “1 Note that this is in line with the common assumption that anomalies are not concentrated (Schölkopf & Smola, 2002; Steinwart et al., 2005).”