Sequential clustering with particle filters-estimating the number of clusters from data

AbstractA basic problem of cluster analysis is the determination or selection of the number of clusters evinced in any set of data. We address this issue with multinomial data using Akaike’s information criterion and demonstrate its utility in identifying an appropriate number of clusters of tumor types with similar profiles of cell surface antigens.

Download Full-text

Standard and Gaussian Particle Filters for Nonlinear System with Missing Measurements

2020 39th Chinese Control Conference (CCC) ◽

10.23919/ccc50068.2020.9188919 ◽

2020 ◽

Author(s):

Xing Zhang ◽

Zhibin Yan

Keyword(s):

Nonlinear System ◽

Particle Filters ◽

Missing Measurements

Download Full-text

HEXACO personality factors and their associations with Facebook use and Facebook network characteristics

10.31234/osf.io/3zvhq ◽

2018 ◽

Author(s):

Riana Brown ◽

Sam G. B. Roberts ◽

Thomas V. Pollet

Keyword(s):

Social Networks ◽

Network Size ◽

Openness To Experience ◽

Personality Factors ◽

Number Of Clusters ◽

Network Characteristics ◽

Online Networks ◽

Facebook Use ◽

Objectively Measured ◽

Size And Structure

Personality factors affect the properties of ‘offline’ social networks, but how they are associated with the structural properties of online networks is still unclear. We investigated how the six HEXACO personality factors (Honesty-Humility, Emotionality, Extraversion, Agreeableness, Conscientiousness and Openness to Experience) relate to Facebook use and three objectively measured Facebook network characteristics - network size, density, and number of clusters. Participants (n = 107, mean age = 20.6, 66% female) extracted their Facebook networks using the GetNet app, completed the 60-item HEXACO questionnaire and the Facebook Usage Questionnaire. Users high in Openness to Experience spent less time on Facebook. Extraversion was positively associated with network size and the number of network clusters (but not after controlling for size). These findings suggest that personality factors are associated with Facebook use and the size and structure of Facebook networks, and that personality is an important influence on both online and offline sociality.

Download Full-text

Effect of leakage and blockage on the acoustic behavior of particle filters

Noise Control Engineering Journal ◽

10.3397/1/376744 ◽

2019 ◽

Vol 67 (6) ◽

pp. 483-492

Author(s):

Seonghyeon Baek ◽

Iljae Lee

Keyword(s):

Transfer Matrix ◽

Particle Filters ◽

Acoustic Analysis ◽

Transmission Loss ◽

One Dimensional ◽

Filter System ◽

The Difference ◽

The One ◽

Analytical Approaches ◽

Good Agreement

The effects of leakage and blockage on the acoustic performance of particle filters have been examined by using one-dimensional acoustic analysis and experimental methods. First, the transfer matrix of a filter system connected to inlet and outlet pipes with conical sections is measured using a two-load method. Then, the transfer matrix of a particle filter only is extracted from the experiments by applying inverse matrices of the conical sections. In the analytical approaches, the one-dimensional acoustic model for the leakage between the filter and the housing is developed. The predicted transmission loss shows a good agreement with the experimental results. Compared to the baseline, the leakage between the filter and housing increases transmission loss at a certain frequency and its harmonics. In addition, the transmission loss for the system with a partially blocked filter is measured. The blockage of the filter also increases the transmission loss at higher frequencies. For the simplicity of experiments to identify the leakage and blockage, the reflection coefficients at the inlet of the filter system have been measured using two different downstream conditions: open pipe and highly absorptive terminations. The experiments show that with highly absorptive terminations, it is easier to see the difference between the baseline and the defects.

Download Full-text

Method for determining optimal number of clusters in K-means clustering algorithm

Journal of Computer Applications ◽

10.3724/sp.j.1087.2010.01995 ◽

2010 ◽

Vol 30 (8) ◽

pp. 1995-1998 ◽

Cited By ~ 18

Author(s):

Shi-bing ZHOU ◽

Zhen-yuan XU ◽

Xu-qing TANG

Keyword(s):

Clustering Algorithm ◽

Optimal Number ◽

Number Of Clusters ◽

Optimal Number Of Clusters

Download Full-text

Clustering Count-based RNA Methylation Data Using a Nonparametric Generative Model

Current Bioinformatics ◽

10.2174/1574893613666180601080008 ◽

2018 ◽

Vol 14 (1) ◽

pp. 11-23 ◽

Cited By ~ 3

Author(s):

Lin Zhang ◽

Yanling He ◽

Huaizhi Wang ◽

Hui Liu ◽

Yufei Huang ◽

...

Keyword(s):

Clustering Analysis ◽

Methylation Level ◽

Optimal Number ◽

Generative Model ◽

Methylation Data ◽

Sequencing Data ◽

Number Of Clusters ◽

Rna Methylation ◽

Clustering Effect ◽

Optimal Number Of Clusters

Background: RNA methylome has been discovered as an important layer of gene regulation and can be profiled directly with count-based measurements from high-throughput sequencing data. Although the detailed regulatory circuit of the epitranscriptome remains uncharted, clustering effect in methylation status among different RNA methylation sites can be identified from transcriptome-wide RNA methylation profiles and may reflect the epitranscriptomic regulation. Count-based RNA methylation sequencing data has unique features, such as low reads coverage, which calls for novel clustering approaches. Objective: Besides the low reads coverage, it is also necessary to keep the integer property to approach clustering analysis of count-based RNA methylation sequencing data. Method: We proposed a nonparametric generative model together with its Gibbs sampling solution for clustering analysis. The proposed approach implements a beta-binomial mixture model to capture the clustering effect in methylation level with the original count-based measurements rather than an estimated continuous methylation level. Besides, it adopts a nonparametric Dirichlet process to automatically determine an optimal number of clusters so as to avoid the common model selection problem in clustering analysis. Results: When tested on the simulated system, the method demonstrated improved clustering performance over hierarchical clustering, K-means, MClust, NMF and EMclust. It also revealed on real dataset two novel RNA N6-methyladenosine (m6A) co-methylation patterns that may be induced directly by METTL14 and WTAP, which are two known regulatory components of the RNA m6A methyltransferase complex. Conclusion: Our proposed DPBBM method not only properly handles the count-based measurements of RNA methylation data from sites of very low reads coverage, but also learns an optimal number of clusters adaptively from the data analyzed. Availability: The source code and documents of DPBBM R package are freely available through the Comprehensive R Archive Network (CRAN): https://cran.r-project.org/web/packages/DPBBM/.

Download Full-text