Outlier Detection Using Replicator Neural Networks
Lecture notes in computer sciencePublished 1 January 2002
Simon Hawkins, Hongxing He, Graham Williams, Rohan A. Baxter
Citations711
SJR quartileQ2
SJR score0.35
SNIP0.55
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
Replicator neural networks (RNNs) are used to provide a measure of the outlyingness of data records and the effectiveness of the RNNs for outlier detection is demonstrated on two publicly available databases.
Abstract
We consider the problem of finding outliers in large multivariate databases. Outlier detection can be applied during the data cleansing process of data mining to identify problems with the data itself, and to fraud detection where groups of outliers are often of particular interest. We use replicator neural networks (RNNs) to provide a measure of the outlyingness of data records. The performance of the RNNs is assessed using a ranked score measure. The effectiveness of the RNNs for outlier detection is demonstrated on two publicly available databases.
Keywords
Computer ScienceMathematics
A density-based algorithm for discovering clusters in large spatial Databases with Noise
19,114 Citations1996Martin Ester, Hans‐Peter Kriegel +2 more
DBSCAN, a new clustering algorithm relying on a density-based notion of clusters which is designed to discover clusters of arbitrary shape, is presented which requires only one input parameter and supports the user in determining an appropriate value for it.
BIRCH
3,906 Citations1996Tian Zhang, Raghu Ramakrishnan +1 more
A data clustering method named BIRCH (Balanced Iterative Reducing and Clustering using Hierarchies) is presented, and it is demonstrated that it is especially suitable for very large databases.
LOF
3,862 Citations2000Markus Breunig, Hans‐Peter Kriegel +2 more
This paper contends that for many scenarios, it is more meaningful to assign to each object a degree of being an outlier, called the local outlier factor (LOF), and gives a detailed formal analysis showing that LOF enjoys many desirable properties.
Cognitive ScienceA Learning Algorithm for Boltzmann Machines*
3,324 Citations1985David H. Ackley, Geoffrey E. Hinton +1 more
Identification of Outliers
2,808 Citations1980D. M. Hawkins
A computer normalizes the one or more sets of historical data points and creates a first visual representation corresponding to the first set of the oneor more sets and the second set of additional points.
ACM SIGMOD RecordEfficient algorithms for mining outliers from large data sets
1,871 Citations2000Sridhar Ramaswamy, Rajeev Rastogi +1 more
Efficient and Effective Clustering Methods for Spatial Data Mining
1,788 Citations1994Raymond T. Ng, Jiawei Han
The analysis and experiments show that with the assistance of CLAHANS, these two algorithms are very effective and can lead to discoveries that are difficult to find with current spatial data mining algorithms.
Algorithms for Mining Distance-Based Outliers in Large Datasets
1,530 Citations1998Edwin M. Knorr, Raymond T. Ng
This paper provides formal and empirical evidence showing the usefulness of DB-outliers and presents two simple algorithms for computing such outliers, both having a complexity of O(k N’), k being the dimensionality and N being the number of objects in the dataset.
The VLDB JournalDistance-based outliers: algorithms and applications
1,196 Citations2000Edwin M. Knorr, Raymond T. Ng +1 more
Outlier detection can be done efficiently for large datasets, and for k-dimensional datasets with large values of k, and it is shown that outlier detection is a meaningful and important knowledge discovery task.
Efficient algorithms for mining outliers from large data sets
1,150 Citations2000Sridhar Ramaswamy, Rajeev Rastogi +1 more
A novel formulation for distance-based outliers that is based on the distance of a point from its kth nearest neighbor is proposed and the top n points in this ranking are declared to be outliers.
Data Mining and Knowledge DiscoveryAdaptive Fraud Detection
873 Citations1997Tom Fawcett, Foster Provost
This paper uses a rule-learning program to uncover indicators of fraudulent behavior from a large database of customer transactions, which are used to create a set of monitors, which profile legitimate customer behavior and indicate anomalies.
Cognitive ScienceA learning algorithm for boltzmann machines
709 Citations1985David H. Ackley, Geoffrey E. Hinton +1 more
Journal of the American Statistical AssociationFast Very Robust Methods for the Detection of Multiple Outliers
255 Citations1994Anthony C. Atkinson
A few repeats of a simple forward search from a random starting point are shown to provide sufficiently robust parameter estimates to reveal masked multiple outliers and stability of the patterns obtained is exhibited by the stalactite plot.
On-line unsupervised outlier detection using finite mixtures with discounting learning algorithms
221 Citations2000Kenji Yamanishi, Jun’ichi Takeuchi +2 more
ScienceReplicator Neural Networks for Universal Optimal Source Coding
135 Citations1995Robert Hecht-Nielsen
A theorem shows that a class of replicator networks can, through the minimization of mean squared reconstruction error, carry out optimal data compression for arbitrary data vector sources.
Journal of the American Statistical AssociationFast Very Robust Methods for the Detection of Multiple Outliers
80 Citations1994Anthony C. Atkinson
Lecture notes in computer scienceMining the knowledge mine
73 Citations1997Graham Williams, Zhexue Huang
The hot spots methodology is presented, adopting a multi-strategy and interactive approach to help focus on the important nuggets of knowledge in insurance and fraud applications.
Computational Statistics & Data AnalysisA procedure for the detection of multivariate outliers
62 Citations1998Andrzej S. Kosinski
A new method for detection of outliers which is very resistant to such high contamination of data with outliers is proposed and detects outliers better than the Rocke and Woodruff method for data highly contaminated (35–45%) with outlier.
A unified approach for mining outliers
53 Citations1997Edwin M. Knorr, Raymond T. Ng
The proposed, intuitive notion of outliers can unify or generalize many of the existing notions of outlier provided by discordancy tests for standard statistical distributions, so that when mining large datasets containing many attributes, a unified approach can replace many statistical discordancies tests, regardless of any knowledge about the underlying distribution of the attributes.
A fast computer intrusion detection algorithm based on hypothesis testing of command transition probabilities
44 Citations1998William DuMouchel, Matthias Schonlau
This statistical method compares in real time the sequence of commands given by each user to a profile of that user's past behavior, using the Fisher score statistic to test the null hypothesis that the observed command transition probabilities come from a profiled transition matrix.
An efficient approximation scheme for data mining tasks
24 Citations2002George Kollios, D. Gunupulos +2 more
A general technique for density-biased sampling that can factor in user requirements to sample for properties of interest, and can be tuned for specific data mining tasks, allowing great flexibility and improved accuracy of the results over simple random sampling is proposed.
Equivalent error bars for neural network classifiers trained by Bayesian inference.
17 Citations1997Peter Sykacek
It is shown that marginalization is not a good method to get moderated probabilities for classes in outlying regions and an alternative measure, that is a more reliable indicator for outliers, is proposed.
Lecture notes in computer scienceThe Integrated Delivery of Large-Scale Data Mining: The ACSys Data Mining Project
9 Citations2002Graham Williams, Irfan Altas +7 more
An overview of the work of the ACSys Data Mining projects is presented where the use of large-scale, high performance computers plays a key role, and the Data Miner's Arcade provides simple abstractions to integrate these components providing high performance data access for a variety of data mining tools communicating through XML.