Choosing DBSCAN Parameters Automatically using Differential Evolution

Karami, Amin; Johansson, Ronnie

doi:10.5120/15890-5059

Cited by 104 publications

(61 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While there are implementations of DBSCAN that automate parameter optimization, these increase time complexity (Ankerst et al, 1999, Karami andJohansson, 2014). As an alternative to automatic parameter estimation, CE allows for the Eps and MinPts parameters to be adjusted via the graphical user interface while viewing the resulting colony identification accuracy.…”

Section: Design and Implementationmentioning

confidence: 99%

Context-explorer: Analysis of spatially organized protein expression in high-throughput screens

Östblom

Nazareth

Tewary

et al. 2018

Preprint

View full text Add to dashboard Cite

A growing body of evidence highlights the importance of the cellular microenvironment as a regulator of phenotypic and functional cellular responses to perturbations. We have previously developed cell patterning techniques to control population context parameters, and here we demonstrate Context-explorer (CE), a software tool to improve investigation of microenvironmental variables through colony level analyses. We demonstrate the capabilities of CE in the analysis of human and mouse pluripotent stem cells (hPSCs, mPSCs) patterned in colonies of defined size and shape in multi-well plates.CE employs a density-based clustering algorithm to identify cell colonies within micropatterned wells. Using this automatic colony classification methodology, we obtain accuracies comparable to manual colony counts in a fraction of the time. Classifying cells according to their relative position within a colony enables statistical analysis of radial spatial trends in protein expression within multiple colonies in the same treatment group. When applied to colonies of hPSCs, our analysis reveals a radial gradient in the expression of the pluripotency inducing transcription factors SOX2 and OCT4, and a similar trend in the intra-colony location of different cellular phenotypes. We extend these analyses to colonies of different sizes and shapes and demonstrate how the metrics derived by CE can be used to asses the patterning fidelity of micropatterned plates.We have incorporated a number of features to enhance the usability and utility of CE. To appeal to a broad scientific community, all of the software's functionality is accessible from a graphical user interface, and convenience functions for several common data operations 2 of 15 May 15, 2018Manuscript Context-explorer are included. CE is compatible with existing image analysis programs such as CellProfiler and extends the analytical capabilities already provided by these tools. Taken together, CE facilitates investigation of spatially heterogeneous cell populations in fundamental research and drug development validation programs.

show abstract

Section: Design and Implementationmentioning

confidence: 99%

Context-explorer: Analysis of spatially organized protein expression in high-throughput screens

Östblom

Nazareth

Tewary

et al. 2018

Preprint

View full text Add to dashboard Cite

show abstract

“…There has been work on automatic parameter selection strategies for density-based clustering, e.g., [8], [17], [22], which are loosely related to the issue illustrated in Figure 1. However, those proposals are unsuitable to be used with HDBSCAN*, since they were developed for non-hierarchical clustering algorithms.…”

Section: Related Workmentioning

confidence: 99%

Efficient Computation of Multiple Density-Based Clustering Hierarchies

Neto

Campello

Nascimento

2017

2017 IEEE International Conference on Data Mining (ICDM)

View full text Add to dashboard Cite

Abstract-HDBSCAN*, a state-of-the-art density-based hierarchical clustering method, produces a hierarchical organization of clusters in a dataset w.r.t. a parameter mpts. While the performance of HDBSCAN* is robust w.r.t. mpts in the sense that a small change in mpts typically leads to only a small or no change in the clustering structure, choosing a "good" mpts value can be challenging: depending on the data distribution, a high or low value for mpts may be more appropriate, and certain data clusters may reveal themselves at different values of mpts.To explore results for a range of mpts values, however, one has to run HDBSCAN* for each value in the range independently, which is computationally inefficient. In this paper, we propose an efficient approach to compute all HDBSCAN* hierarchies for a range of mpts values by replacing the graph used by HDBSCAN* with a much smaller graph that is guaranteed to contain the required information. An extensive experimental evaluation shows that with our approach one can obtain over one hundred hierarchies for the computational cost equivalent to running HDBSCAN* about 2 times.

show abstract

“…In order to find the most appropriate parameters, an optimization algorithm can be used [38,39]. An optimization algorithm will attempt to find an optimal choice that satisfies defined constraints and make an optimization criterion (performance or cost index) maximize or minimize [38,40]. Hence, to improve the prediction accuracy and robustness of the RBF network, network parameters (centers, widths and weights) should be simultaneously tuned [32].…”

Section: Introductionmentioning

confidence: 99%

A hybrid multiobjective RBF-PSO method for mitigating DoS attacks in Named Data Networking

Karami

Zapata

2015

Neurocomputing

Self Cite

View full text Add to dashboard Cite

Named Data Networking (NDN) is a promising network architecture being considered as a possible replacement for the current IP-based (host-centric) Internet infrastructure. NDN can overcome the fundamental limitations of the current Internet, in particular, Denial-of-Service (DoS) attacks. However, NDN can be subject to new type of DoS attacks namely Interest flooding attacks and content poisoning. These types of attacks exploit key architectural features of NDN. This paper presents a new intelligent hybrid algorithm for proactive detection of DoS attacks and adaptive mitigation reaction in NDN. In the detection phase, a combination of multiobjective evolutionary optimization algorithm with PSO in the context of the RBF neural network has been applied in order to improve the accuracy of DoS attack prediction. Performance of the proposed hybrid approach is also evaluated successfully by some benchmark problems. In the adaptive reaction phase, we introduced a framework for mitigating DoS attacks based on the misbehaving type of network nodes. The evaluation through simulations shows that the proposed intelligent hybrid algorithm (proactive detection and adaptive reaction) can quickly and effectively respond and mitigate DoS attacks in adverse conditions in terms of the applied performance criteria.

show abstract

Choosing DBSCAN Parameters Automatically using Differential Evolution

Cited by 104 publications

References 50 publications

Context-explorer: Analysis of spatially organized protein expression in high-throughput screens

Context-explorer: Analysis of spatially organized protein expression in high-throughput screens

Efficient Computation of Multiple Density-Based Clustering Hierarchies

A hybrid multiobjective RBF-PSO method for mitigating DoS attacks in Named Data Networking

Contact Info

Product

Resources

About