Interpreting and Unifying Outlier Scores

By Hans-Peter Kriegel, Peer Kröger, Erich Schubert, Arthur Zimek. SDM, 2011

300 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/a48a8b2085262a7f9658368088eca756e400d5ae

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

In Statistics and Probability

Citations in the corpus

Outlier Analysis (2013, methodology): “This approach lies at the core of ensemble-based methods [289, 310, 271].”
Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection (2015, methodology): “However, for densitybased methods, a horizontal cut corresponds to a single, global density threshold which may not simultaneously detect clusters having largely varying local densities [Ankerst et al. 1999; Tan et al. 2006; Kriegel et al. 2011].”
A survey on unsupervised outlier detection in high‐dimensional numerical data (2012, methodology): “The normalization [116] and the diversity analysis demonstrated in ref.”
HiCS: High Contrast Subspaces for Density-Based Outlier Ranking (2012, background): “The implementation details are given in Section IV-A.”
Deep Learning for Anomaly Detection (2020, methodology): “h multiple output heads in [109]; Eq. (39) presents a simplifiedτfor brevity. The idea of enforcing a prior on the anomaly scores is explored in [116]. Motivated by the extensive empirical results in [75] that show the anomaly scores in a variety of real-world data sets fits Gaussian distribution ve”
Progress in Outlier Detection Techniques: A Survey (2019, methodology): “Other succeeding studies [38], [190], [191] in later years that focused on using ensembles for outlier detection faced numerous challenges.”
KATARA: A Data Cleaning System Powered by Knowledge Bases and Crowdsourcing (2015, background): “tion from scores to probabilities follows the general framework of interpreting scores in [25].”
Deep Anomaly Detection with Deviation Networks (2019, background): “Additionally, in contrast to most methods that produce hardly interpretable anomaly scores [10], the Z-Score-based deviation loss also allows DevNet to produce easily interpretable anomaly scores.”
Contextual Outlier Interpretation (2017, background): “Besides non-negativeness and niteness, an important requirement for an outlierness measure is ranking-stability [23].”
Iterative Learning with Open-set Noisy Labels (2018, background): “1As pointed by [17, 18], pcLOF is not sensitive to parameter k Figure 4.”
An instance level analysis of data complexity (2014, background): “There are a number of approaches that aim at overcoming the uninterpretability of LOF (Kriegel et al. 2011).”
PyOD: A Python Toolbox for Scalable Outlier Detection (2019, methodology): “Inspired by scikit-learn ’s API design, all implemented outlier detection algorithms inherit from a base class with the same interface: (i) fit processes the train data and computes the necessary statistics; (ii) decision function generates raw outlier scores for unseen data after the model is ﬁtted”