LOF: identifying density-based local outliers
By M. Breunig, Hans-Peter Kriegel, Raymond T. Ng, Jörg Sander. 2000
8,778 citations recorded by the source corpus.
For many KDD applications, such as detecting criminal activities in E-commerce, finding the rare instances or the outliers, can be more interesting than finding the common patterns. Existing work in outlier detection regards being an outlier as a binary property. In this paper, we contend that for many scenarios, it is more meaningful to assign to each object a degree of being an outlier. This degree is called the local outlier factor (LOF) of an object. It is local in that the degree depends on how isolated the object is with respect to the surrounding neighborhood. We give a detailed formal analysis showing that LOF enjoys many desirable properties. Using real-world datasets, we demonstrate that LOF can be used to find outliers which appear to be meaningful, but can otherwise not be identified with existing approaches. Finally, a careful performance evaluation of our algorithm confirms we show that our approach of finding local outliers can be practical.
Subfields: Artificial Intelligence, Biomedical Engineering, Signal Processing
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
- Charu C. Aggarwal
- Yoshua Bengio
- Percy Liang
- Dacheng Tao
- Mohit Bansal
- Quoc V. Le
- Jianfeng Gao
- Sergey Levine
In Biomedical Engineering
- Charu C. Aggarwal
- Arthur Zimek
- Erich Schubert
- Hans-Peter Kriegel
- Markus Goldstein
- Ricardo José Gabrielli Barreto Campello
- Jörg Sander
- Rfdo 2XWOLHUV
In Signal Processing
- Charu C. Aggarwal
- Arthur Zimek
- Jiawei Han
- Vipin Kumar
- Hans-Peter Kriegel
- Markus Goldstein
- Jörg Sander
- Aleksandar Lazarevic
Citations in the corpus
- Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift (2018, methodology): “Popular approaches to anomaly detection include density estimation [6], margin-based approaches such as the one-class SVM [40], and the tree-based isolation forest method due to [30].”
- Scaling Out-of-Distribution Detection for Real-World Settings (2022, methodology): “has a rich history spanning several decades [34, 5, 11].”
- Timeseries Anomaly Detection using Temporal Hierarchical One-Class Network (2020, neutral): “These include traditional one-class classifiers: (i) local outlier factor (LOF) [1], (ii) one-class SVM (OC-SVM) with RBF kernel [29], and (iii) isolation forest [16]), and the recent deep learning models of (i) deep SVDD [22], (ii) AnoGAN [24], and (iii) deep autoencoding Gaussian mixture model…”
- Interpreting and Unifying Outlier Scores (2011, background): “The basic local outlier factor (LOF) assigned to each object of the database D denotes a degree of outlierness [9].”
- TensorCast: forecasting and mining with coupled tensors (2018, neutral): “For a cloud of points, an anomaly is a point that is ‘too far away’ from the rest [8]—this is exactly what outlier detection methods do (like the time-tested LOF [7], LOCI [27], and many more, up to the very recent Random-Cut-Forest-based anomaly detection [14]).”
- Outlier Detection with Autoencoder Ensembles (2017, methodology): “Numerous methods such as distance-based methods [5, 11, 17], density-based methods [6], linear methods [21], and spectral methods [8, 18] have been proposed.”
- Passban IDS: An Intelligent Anomaly-Based Intrusion Detection System for IoT Edge Devices (2020, methodology): “2) Local Outlier Factor: It is a density-based method in which the distance between given data points is measured to estimate their density in a locality [55].”
- ADBench: Anomaly Detection Benchmark (2022, background): “Definition and Generation Process of Four Types of Common Anomalies Used in ADBench : • Local anomalies refer to the anomalies that are deviant from their local neighborhoods [22].”
- Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding (2018, methodology): “d of other anomaly detection techniques have been introduced and explored as potential improvements over OOL approaches, such as clustering-based approaches [15, 24, 28], nearest neighbors approaches [3, 6, 23, 25], expert systems [7, 34, 36, 43], and dimensionality reduction approaches [14, 39, 45]”
- Generalized Out-of-Distribution Detection: A Survey (2021, neutral): “Density-Based Methods OD methods include Gaussian distribution (Altman & Bland, 2005; Leys et al., 2013), Mahalanobis distance (De Maesschalck et al., 2000), Gaussian mixtures (Yang et al., 2009), and Local outlier factor (LOF) (Breunig et al., 2000).”
- Knowledge graph refinement: A survey of approaches and evaluation methods (2016, methodology): “In that feature space, standard outlier detection techniques such as Local Outlier Factor [11] or cluster-based outlier detection [34] can be used to assign outlier scores.”
- Isolation-Based Anomaly Detection (2012, methodology): “The four state-of-the-art anomaly detectors used in experiments are ORCA [Bay and Schwabacher 2003], one-class SVM [Sch¨olkopf et al. 2001], LOF [Breunig et al. 2000] and Random Forests (RF) [Shi and Horvath 2006].”