• The Korean Journal of Applied Statistics
• /
• v.33 no.4
• /
• pp.511-526
• /
• 2020

Single-cell RNA-sequencing (scRNA-seq) data consists of each cell's RNA expression extracted from large populations of cells. One main purpose of using scRNA-seq data is to identify inter-cellular heterogeneity. However, scRNA-seq data pose statistical challenges when applying traditional clustering methods because they have many missing values and high level of noise due to technical and sampling issues. In this paper, motivated by analyzing scRNA-seq data, we propose a novel spectral-based clustering method by imposing different weights on genes when computing a similarity between cells. Assigning weights on genes and clustering cells are performed simultaneously in the proposed clustering framework. We solve the proposed non-convex optimization using an iterative algorithm. Both real data application and simulation study suggest that the proposed clustering method better identifies underlying clusters compared with existing clustering methods.

• International conference on construction engineering and project management
• /
• 2015.10a
• /
• pp.559-563
• /
• 2015

Identification of energy use patterns in buildings has a great opportunity for energy saving. To find what energy use patterns exist, clustering analysis has been commonly used such as K-means and hierarchical clustering method. In case of high dimensional data such as energy use time-series, data reduction should be considered to avoid the curse of dimensionality. Principle Component Analysis, Autocorrelation Function, Discrete Fourier Transform and Discrete Wavelet Transform have been widely used to map the original data into the lower dimensional spaces. However, there still remains an ongoing issue since the performance of clustering analysis is dependent on data type, purpose and application. Therefore, we need to understand which data reduction techniques are suitable for energy use management. This research aims find the best clustering method using energy use data obtained from Seoul National University campus. The results of this research show that most experiments with data reduction techniques have a better performance. Also, the results obtained helps facility managers optimally control energy systems such as HVAC to reduce energy use in buildings.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.2
• /
• pp.395-406
• /
• 2017

The purpose of this study is to identify the pattern of daily electricity demand through clustering and classification. The hourly data was collected by KPS (Korea Power Exchange) between 2008 and 2012. The time trend was eliminated for conducting the pattern of daily electricity demand because electricity demand data is times series data. We have considered k-means clustering, Gaussian mixture model clustering, and functional clustering in order to find the optimal clustering method. The classification analysis was conducted to understand the relationship between external factors, day of the week, holiday, and weather. Data was divided into training data and test data. Training data consisted of external factors and clustered number between 2008 and 2011. Test data was daily data of external factors in 2012. Decision tree, random forest, Support vector machine, and Naive Bayes were used. As a result, Gaussian model based clustering and random forest showed the best prediction performance when the number of cluster was 8.

• Journal of Korea Multimedia Society
• /
• v.21 no.8
• /
• pp.952-959
• /
• 2018

Clustering is one of the most fundamental algorithms in machine learning. The performance of clustering is affected by the distribution of data, and when there are more data or more dimensions, the performance is degraded. For this reason, we use a stacked auto encoder, one of the deep learning algorithms, to reduce the dimension of data which generate a feature vector that best represents the input data. We use k-means, which is a famous algorithm, as a clustering. Sine the feature vector which reduced dimensions are also multi dimensional, we use the Euclidean distance as well as the cosine similarity to increase the performance which calculating the similarity between the center of the cluster and the data as a vector. A deep clustering networks combining a stacked auto encoder and k-means re-trains the networks when the k-means result changes. When re-training the networks, the loss function of the stacked auto encoder and the loss function of the k-means are combined to improve the performance and the stability of the network. Experiments of benchmark image ad document dataset empirically validated the power of the proposed algorithm.

• The Korean Journal of Applied Statistics
• /
• v.17 no.1
• /
• pp.135-144
• /
• 2004

We propose a reproducibility (validity) assessment procedure of K-means cluster analysis by randomly partitioning the data set into three parts, of which two subsets are used for developing clustering rules and one subset for testing consistency of clustering rules. Also, as an alternative to Rand index and corrected Rand index, we propose an entropy-based consistency measure between two clustering rules, and apply it to determination of the number of clusters in K-means clustering.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.20 no.3
• /
• pp.105-111
• /
• 2010

We propose a practical privacy-preserving clustering protocol over horizontally partitioned data. We extend the DBSCAN clustering algorithm into a distributed protocol in which data providers mix real data with fake data to provide privacy. Our privacy-preserving clustering protocol is very efficient whereas the previous privacy-preserving protocols in the distributed environments are not practical to be used in real applications. The efficiency of our privacy-preserving clustering protocol over horizontally partitioned data is comparable with those of privacy-preserving clustering protocols in the non-distributed environments.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.17 no.1
• /
• pp.75-84
• /
• 2021

For analyzing big data, the social network is increasingly being utilized through relational data, which means the connection characteristics between entities such as people and objects. When the relational data does not exist directly, a social network can be configured by calculating relational data such as attribute similarity from attribute data of entities and using it as links. In this paper, the composition method of the social network using the attribute similarity between entities as a connection relationship, and the clustering method using subgroups for the configured social network are suggested, and the clustering effectiveness of the clustering results is evaluated. The analysis results can vary depending on the type and characteristics of the data to be analyzed, the type of attribute similarity selected, and the criterion value. In addition, the clustering effectiveness may not be consistent depending on the its evaluation method. Therefore, selections and experiments are necessary for better analysis results. Since the analysis results may be different depending on the type and characteristics of the analysis target, options for clustering, etc., there is a limitation. In addition, for performance evaluation of clustering, a study is needed to compare the method of this paper with the conventional method such as k-means.

• The Korean Journal of Applied Statistics
• /
• v.18 no.3
• /
• pp.661-671
• /
• 2005

The aim of this study is to propose a clustering method, called tree-structured clustering, by recursively partitioning continuous multivariate dat a based on overall $R^2$ criterion with a practical node-splitting decision rule. The clustering method produces easily interpretable clustering rules of tree types with the variable selection function. In numerical examples (Fisher's iris data and a Telecom case), we note several differences between tree-structured clustering and K-means clustering.

• Journal of Information Processing Systems
• /
• v.13 no.6
• /
• pp.1467-1486
• /
• 2017

Fuzzy co-clustering is sensitive to noise data. To overcome this noise sensitivity defect, possibilistic clustering relaxes the constraints in FCM-type fuzzy (co-)clustering. In this paper, we introduce a new possibilistic fuzzy co-clustering algorithm based on information bottleneck (ibPFCC). This algorithm combines fuzzy co-clustering and possibilistic clustering, and formulates an objective function which includes a distance function that employs information bottleneck theory to measure the distance between feature data point and feature cluster centroid. Many experiments were conducted on three datasets and one artificial dataset. Experimental results show that ibPFCC is better than such prominent fuzzy (co-)clustering algorithms as FCM, FCCM, RFCC and FCCI, in terms of accuracy and robustness.

• Genomics & Informatics
• /
• v.10 no.4
• /
• pp.263-265
• /
• 2012

We developed a user-friendly, interactive program to simultaneously cluster and visualize omics data, such as DNA and protein array profiles. This program provides diverse algorithms for the hierarchical clustering of two-dimensional data. The clustering results can be interactively visualized and optimized on a heatmap. The present tool does not require any prior knowledge of scripting languages to carry out the data clustering and visualization. Furthermore, the heatmaps allow the selective display of data points satisfying user-defined criteria. For example, a clustered heatmap of experimental values can be differentially visualized based on statistical values, such as p-values. Including diverse menu-based display options, QCanvas provides a convenient graphical user interface for pattern analysis and visualization with high-quality graphics.