Identification of Outliers

Hawkins, Douglas M.

doi:10.1007/978-94-015-3994-4

Cited by 2,250 publications

(689 citation statements)

References 60 publications

Supporting

Mentioning

649

Contrasting

Unclassified

Order By: Relevance

“…These points deep inside a dense cluster have a LOF value of approximately 1 while the isolated points have a much higher value. The authors claim that this definition also catches the spirit of the outlier definition given by Hawkins [6]. The local outlier notion seems more reasonable than DB-outlier because each data point can be measured with a numerical factor based on how the data is deviated from its genuine cluster.…”

Section: Related Workmentioning

confidence: 86%

“…Hawkins defines an outlier as "an observation which deviates so much from other observations as to arouse suspicions that it was generated by a different mechanism" [6]. Traditionally outlier detection in engineering disciplines depends on statistical approaches.…”

Section: Related Workmentioning

confidence: 99%

“…While considering locality and globality, our definition still embodies the essence of the accepted definition given by Hawkins [6] with regard to deviation from other observations.…”

Section: Definition 2 Resolution-based Outlier Factor (Rof)mentioning

confidence: 99%

See 2 more Smart Citations

A Nonparametric Outlier Detection for Effectively Discovering Top-N Outliers from Engineering Data

Fan

Zaı̈ane

Foss

et al. 2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We present a novel resolution-based outlier notion and a nonparametric outlier-mining algorithm, which can efficiently identify top listed outliers from a wide variety of datasets. The algorithm generates reasonable outlier results by taking both local and global features of a dataset into consideration. Experiments are conducted using both synthetic datasets and a real life construction equipment dataset from a large building contractor. Comparison with the current outlier mining algorithms indicates that the proposed algorithm is more effective.

show abstract

Section: Related Workmentioning

confidence: 86%

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A Nonparametric Outlier Detection for Effectively Discovering Top-N Outliers from Engineering Data

Fan

Zaı̈ane

Foss

et al. 2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Consequently, most existing tests are not usable here because they only work on large datasets (e.g. Rosner's Test [17] and others [5], [13]), and those that were usable for small data sets gave poor results (e.g. Dixon's test [9] and others).…”

Section: Identifying Superfluous Informationmentioning

confidence: 99%

Automatic Runtime Validation and Correction of the Navigational Design of Web Sites

Casteleyn

Garrigós

Troyer

2005

Web Technologies Research and Development - APWeb 2005

View full text Add to dashboard Cite

Abstract. Essential to an audience driven website design philosophy is the organization of information and functionality according to the requirements of the different audience classes. However, at design time, it is often difficult to correctly assess the different needs and requirements of the different prospective users of a website. This may result in a non-optimal navigation structure, which will decrease the usability of the website. In this paper, we describe how to correct, at run-time and automatically, possible flaws in the design resulting from incomplete requirement assessment, using adaptive behavior. By observing the browsing behavior of the users, the requirements for the different users are validated and the website is adapted according to adaptation specifications made by the designer. These specifications express when and how the website needs to be adapted and are expressed using an Adaptation Specification Language. The work is presented in the context of an audience driven design method but we also elaborate shortly on the applicability of the technique in general.

show abstract

“…In data and pattern analysis, large outliers influence the statistics of the analysis and can lead to inconclusive and/or wrong results. There are many methods for detection of outliers, including methods based on statistical data distributions, prior knowledge of the nature of distributions, expected number of outliers, and the nature of expected outliers [1,2]. Majority of these methods are only applicable to univariate data or to a specific type of data distribution.…”

Section: Introductionmentioning

confidence: 99%

Inter-image outliers and their application to image classification

James

Dimitrijev

2010

Pattern Recognition

View full text Add to dashboard Cite

Image variability that is impossible or difficult to restore by intra-image processing, such as the variability caused by occlusions, significantly reduces the performance of image-recognition methods. To address this issue, we propose that the pixels associated with large distances obtained by inter-image pixel-by-pixels comparisons should be considered as inter-image outliers and should be removed from the similarity calculation used for the image classification. When this method is combined with the template-matching method for image recognition, it leads to state-of-the-art recognition performance: 91% with AR database that includes occluded face images, 90% with PUT database that includes pose variations of face images and 100% with EYale B database that includes images with large illumination variation.

show abstract

Identification of Outliers

Cited by 2,250 publications

References 60 publications

A Nonparametric Outlier Detection for Effectively Discovering Top-N Outliers from Engineering Data

A Nonparametric Outlier Detection for Effectively Discovering Top-N Outliers from Engineering Data

Automatic Runtime Validation and Correction of the Navigational Design of Web Sites

Inter-image outliers and their application to image classification

Contact Info

Product

Resources

About