A comparative study of RNN for outlier detection in data mining

By Graham J. Williams, Rohan A. Baxter, Hongxing He, Simon Hawkins, Lifang Gu. IEEE International Conference on Data Mining. Proceedings, 2002

308 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/390132e92cd34b912c56fd53dc086d53abb8780e

How researchers see this paper

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

In Artificial Intelligence

Citations in the corpus

Fake Twitter followers detection by denoising autoencoder (2017, methodology): “In [17] and [27] an autoencoder with three hidden layers and different activation functions based on hyperbolic tangent has been used.”
A Comprehensive Survey of Data Mining-based Fraud Detection Research (2010, methodology): “Unsupervised approaches include Hawkins et al (2002) and Williams et al (2002) which advocate replicator neural networks to detect outliers.”
Outlier Detection Using k-Nearest Neighbour Graph (2004, background): “Similarly computer security intrusion detection system finds an inlier pattern as representation of normal network behaviour and outliers as possible intrusion attempts [13].”
Outlier Detection with Autoencoder Ensembles (2017, background): “Comparisons of neural network methods may be found in [22] and a detailed survey on neural networks may be found in [13].”
Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection (2018, methodology): “Under this framework, a discriminative boundary surrounding the normal instances is learned by algorithms, such as one-class SVM (Chen et al. (2001); Song et al. (2002); Williams et al. (2002)).”
Deep Structured Energy Based Models for Anomaly Detection (2016, methodology): “…(KDE) and the more recent robust kernel density estimator (RKDE); (3) methods of learning a compact data model such that as many as possible normal samples are enclosed inside, for example, one-class SVM and SVDD. Graham et al. proposed a method based on autoencoder (Williams et al., 2002).”
Anomaly detection: A survey (2009, methodology): “Replicator Neural Networks have been used for one-class anomaly detection[Hawkins et al. 2002; Williams et al. 2002].”
Anomaly Detection Using Autoencoders with Nonlinear Dimensionality Reduction (2014, background): “Although some works compare an autoencoder with other approaches [7], [15], in this paper we focus on the dimensionality reduction and determine the difference in performances according to the reconstruction error.”
Outlier Analysis (2013, methodology): “As shown in [209] direct adaptations of various methods such as replicator neural networks [464], and clustering methods [210] do not work effectively.”
Isolation-Based Anomaly Detection (2012, methodology): “…anomaly detection approaches, including classi.cation-based methods [Abe et al. 2006; Shi and Horvath 2006], Replicator Neural Network (RNN) [Williams et al. 2002], one-class SVM [Tax and Duin 2004] and clustering-based methods [He et al. 2003], construct a pro.le of normal instances,…”