Cluster validity measure and merging system for hierarchical clustering considering outliers

Morsier, Frank de; Tuia, Devis; Borgeaud, M.; Gass, Volker; Thiran, Jean-Philippe

doi:10.1016/j.patcog.2014.10.003

Cited by 27 publications

(12 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The tree is called a Dendrogram. In this research, the single-link divisive clustering algorithm [23] is employed. It is one of the oldest and simplest clustering methods, known as the nearest neighbor method.…”

Section: Hierarchical Clusteringmentioning

confidence: 99%

A Comparison among Data Mining Algorithms for Outlier Detection using Flow Pattern Experiments

2017

View full text Add to dashboard Cite

Abstract. Accurate outlier detection is an important matter to consider prior to applying data to predict ow patterns. Identifying these outliers and reducing their impact on measurements could be e ective in presenting an authentic ow pattern. This paper aims to detect outliers in ow pattern experiments along a 180-degree sharp bend channel with and without a T-shaped spur dike. Velocity components have been collected using 3D velocimeter called Vectrino in order to determine the ow pattern. Some of outlier detection methods were employed in the paper, such as Z-score test, sum of sine curve tting, Mahalanobis distance, hierarchical clustering, LSC-mine, self-organizing map, fuzzy C-means clustering, and voting. Considering the experiments carried out, the methods were e cient in outlier detection; however, the voting method appeared to be the most e cient one. Brie y, this paper calculated di erent hydraulic parameters in the sharp bend and made a comparison between them for the sake of studying how e ective running the voting method is in mean and turbulent ow pattern variations. The results indicated that developing the voting method in the ow pattern experiment in the bend would cause a decrease in Reynolds shear stress by 36%, while the mean velocities were not signi cantly in uenced by the method.

show abstract

Section: Hierarchical Clusteringmentioning

confidence: 99%

A Comparison among Data Mining Algorithms for Outlier Detection using Flow Pattern Experiments

2017

View full text Add to dashboard Cite

show abstract

“…Starczewski (2015)) has been defined as the product of two components which determine changes of compactness and separability of clusters during a clustering process. De Morsier et al, (2015) quantify the clustering performance of hierarchical algorithms that handle overlapping clusters in presence of outliers. A set of validation techniques based on the clusters' negentropy is discussed in (Lago-Fernandez et al, (2014)).…”

Section: Literature Reviewmentioning

confidence: 99%

A new cluster validity index using maximum cluster spread based compactness measure

Wani

Riyaz

2016

International Journal of Intelligent Computing and Cybernetics

View full text Add to dashboard Cite

show abstract

“…Specifically, the probability density function (pdf) of the data provided by annotators is modeled as a mixture of an unknown number of Gaussian components plus a uniformly distributed random variable (rv), which models the annotators' isolated errors as outliers. Unlike previous works for clustering, e.g., [14,15,16,17], the proposed EM-based clustering algorithm not only estimates the number of Gaussian components and the parameters of the Gaussian plus non-Gaussian mixture density, but also annotators' reliability. In the detection stage, a decision is made, on each cluster identified in the clustering step, on whether it corresponds to one of the desired structures or not, taking into account annotators' reliability.…”

Section: Introductionmentioning

confidence: 99%

Unsupervised online clustering and detection algorithms using crowdsourced data for malaria diagnosis

Pagès-Zamora

Cabrera-Bean

Diaz-Vilor

2019

Pattern Recognition

View full text Add to dashboard Cite

Crowdsourced data in science might be severely error-prone due to the inexperience of annotators participating in the project. In this work, we present a procedure to detect specific structures in an image given tags provided by multiple annotators and collected through a crowdsourcing methodology. The procedure consists of two stages based on the Expectation-Maximization (EM) algorithm, one for clustering and the other one for detection, and it gracefully combines data coming from annotators with unknown reliability in an unsupervised manner. An online implementation of the approach is also presented that is well suited to crowdsourced streaming data. Comprehensive experimental results with real data from the MalariaSpot project are also included.

show abstract

Cluster validity measure and merging system for hierarchical clustering considering outliers

Cited by 27 publications

References 55 publications

A Comparison among Data Mining Algorithms for Outlier Detection using Flow Pattern Experiments

A Comparison among Data Mining Algorithms for Outlier Detection using Flow Pattern Experiments

A new cluster validity index using maximum cluster spread based compactness measure

Unsupervised online clustering and detection algorithms using crowdsourced data for malaria diagnosis

Contact Info

Product

Resources

About