LOCI: fast outlier detection using the local correlation integral

By Spiros E. Papadimitriou, H. Kitagawa, Phillip B. Gibbons, Christos Faloutsos. Proceedings / International Conference on Data Engineering, 2003

1,024 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Networks and Communications, Signal Processing

View Paper: https://www.semanticscholar.org/paper/88c2c3772201d4ee244e80b05f577cf79b454531

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 Computer Networks and Communications

In Signal Processing

Citations in the corpus

Hierarchical Density Estimates for Data Clustering, Visualization, and Outlier Detection (2015, methodology): “We also compare to LOCI [Papadimitriou et al. 2003] as the only competitor that also tries to find both types of outliers, global and local ones.”
Progress in Outlier Detection Techniques: A Survey (2019, methodology): “Some of the density-based techniques [8], [75], [81], [82] have served as a fundamental baseline for many subsequent algorithms.”
Outlier Analysis (2013, methodology): “It has been shown in [356], that a box count of a grid-based division of the data provides a fast approximation, when L∞ distances are used.”
Ranking Outliers Using Symmetric Neighborhood Relationship (2006, methodology): “LOCI [17] addresses the difficulty of choosing values for MinPts in the LOF technique by using statistical values derived from the data itself.”
ECOD: Unsupervised Outlier Detection Using Empirical Cumulative Distribution Functions (2022, background): “A few improvements were later introduced, including the Connective-based Out-lier Factor (COF) by Tang et al. [35], and LOcal Correlation Integral (LOCI) by Papadimitriou et al. [36].”
PyOD: A Python Toolbox for Scalable Outlier Detection (2019, methodology): “, 2003) Proximity Yes No OCSVM (Ma and Perkins, 2003) Linear Model No No LOCI (Papadimitriou et al., 2003) Proximity Yes No PCA (Shyu et al.”
Graph based anomaly detection and description: a survey (2014, methodology): “The techniques can be classified into density-based [Breunig et al., 2000, Papadimitriou et al., 2003], distance-based [Aggarwal and Yu, 2001, Chaudhary et al., 2002, Ghoting et al., 2008, Knorr and Ng, 1998, Orair et al., 2010, Wang et al., 2011b], depth-based [Ruts and Rousseeuw, 1996],…”
Incremental Local Outlier Detection for Data Streams (2007, background): “In addition, an extension of the proposed methodology to create incremental versions of other emerging outlier detection algorithms, (e.g., Connectivity Outlier Factor (COF) [34], LOCI [35]), is also worth considering.”
A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data (2016, methodology): “For both, LOCI and aLOCI, the recommended parameter settings from the authors were used [47].”
Outlier Detection Techniques for Wireless Sensor Networks: A Survey (2008, methodology): “In the other technique, each node employs more robust technique called LOCI [36] to locally detect global outliers by having a copy of global estimator model obtained from the sink.”
State of research in automatic as-built modelling (2015, methodology): “The method proposed in [88] compares the local sampling density of a point to the average local sampling density of its neighbours, and identifies outliers in a parameterless setup.”
k-means-: A Unified Approach to Clustering and Outlier Detection (2013, background): “This can result in many false negatives: i.e., data points that should be declared outliers are masked by the clustering and also false positives: data points that are incorrectly labeled as outliers.”
oddball: Spotting Anomalies in Weighted Graphs (2010, methodology): “Typical methods include LOF [7] and LOCI [27].”
Anomaly Detection in Time Series: A Comprehensive Evaluation (2022, methodology): “Outlier Detection LOCI/aLOCI [103] MCOD[73] Isolation Forest [83] CBLOF [59]”
LoOP: local outlier probabilities (2009, background): “Traditional approaches such as LOF and LOCI may even emphasize such effects.”
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 Kernel Density Functions (2007, methodology): “Our experiments performed on several simulated data sets have demonstrated that the proposed approach outperforms two very popular density-based outlier detection algorithms, LOF [9] and LOCI [35].”