• Journal of Convergence for Information Technology
• /
• v.12 no.5
• /
• pp.100-106
• /
• 2022

A longitudinal study refers to a research method based on longitudinal data repeatedly measured on the same object. Most of the longitudinal analysis methods are suitable for prediction or inference, and are often not suitable for use in exploratory study. In this study, an exploratory method to analyze longitudinal data is presented, which is to find the longitudinal trajectory after determining the best number of clusters by clustering longitudinal data using self-organizing map technique. The proposed methodology was applied to the longitudinal data of the Employment Information Service, and a total of 2,610 samples were analyzed. As a result of applying the methodology to the actual data applied, time-series clustering results were obtained for each panel. This indicates that it is more effective to cluster longitudinal data in advance and perform multilevel longitudinal analysis.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.24 no.2
• /
• pp.73-84
• /
• 2022

Based on the method of marking the size of children's wear suggested by Korean Agency for Technology and Standards, this study classified the heights of school-aged girls by 10cm. The purpose of this study is to provide foundational material by analyzing and comparing the characteristics of torso types. The author analyzed the height section of 130 to 139.9cm, which the majority of subjects fell into, and concerning the method of research, this study statistically analyzed body sizes and calculated items related to 162 persons' torso types out of the 6th Korean Body Size Data. According to the results, eight factors were extracted, and the total explanatory variate of all the factors was found to be 81.93%. According to the results of cluster analysis with it as an independent variable, three types were drawn. Type 1 (41.4%) was found to be the thickest in the torso and round and severely curved in the sectional form of the circumference item. RegardingType 2 (25.9%), the horizontal size of the torso is similar to that of Type 1: the upper body is long, and the sectional form of the circumference item is oval-shaped. Regarding Type 3 (32.7%), mean values are similar to those of Type 2 overall: the upper body is short, and the body is the most upright. In conclusion, according to the results of analyzing torso types, the types and average values indicate significant differences in the height section of 130~139.9cm. This implies that when making ready-made clothes, it is necessary to come up with the sizes of more detailed items in relation with height. It is expected that the findings of this study will be utilized as basic data when children's wear companies develop prototypes and use grading variances according to the variations of size.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.23-43
• /
• 2022

In order to apply AI (Artificial Intelligence) in various industries, interest in algorithm selection is increasing. Algorithm selection is largely determined by the experience of a data scientist. However, in the case of an inexperienced data scientist, an algorithm is selected through meta-learning based on dataset characteristics. However, since the selection process is a black box, it was not possible to know on what basis the existing algorithm recommendation was derived. Accordingly, this study uses k-means cluster analysis to classify types according to data set characteristics, and to explore suitable classification algorithms and methods for resolving class imbalance. As a result of this study, four types were derived, and an appropriate class imbalance resolution method and classification algorithm were recommended according to the data set type.

• Journal of Veterinary Science
• /
• v.21 no.1
• /
• pp.9.1-9.15
• /
• 2020

Regenerative therapy holds great promise in the development of cures of some untreatable diseases such as cardiovascular diseases, and pluripotent stem cells (PSCs) including induced PSCs (iPSCs) are the most important regenerative seed cells. Recently, differentiation of human PSCs into functional tissues and cells in vitro has been widely reported. However, although porcine reports are rare they are quite essential, as the pig is an important animal model for the in vitro generation of human organs. In this study, we reprogramed porcine embryonic fibroblasts into porcine iPSCs (piPSCs), and differentiated them into cluster of differentiation 31 (CD31)-positive endothelial cells (ECs) (piPSC-derived ECs, piPS-ECs) using an optimized single-layer culture method. During differentiation, we observed that a combination of GSK3β inhibitor (CHIR99021) and bone morphogenetic protein 4 (BMP4) promoted mesodermal differentiation, resulting in higher proportions of CD31-positive cells than those from separate CHIR99021 or BMP4 treatment. Importantly, the piPS-ECs showed comparable morphological and functional properties to immortalized porcine aortic ECs, which are capable of taking up low-density lipoprotein and forming network structures on Matrigel. Our study, which is the first trial on a species other than human and mouse, has provided an optimized single-layer culture method for obtaining ECs from porcine PSCs. Our approach can be beneficial when evaluating autologous EC transplantation in pig models.

• Journal of Digital Convergence
• /
• v.20 no.2
• /
• pp.217-223
• /
• 2022

Subspace clustering for online data streams requires a large amount of memory resources as all subsets of data dimensions must be examined. In order to track the continuous change of clusters for a data stream in a finite memory space, in this paper, we propose a grid-based subspace clustering algorithm that effectively uses memory resources. Given an n-dimensional data stream, the distribution information of data items in data space is monitored by a grid-cell list. When the frequency of data items in the grid-cell list of the first level is high and it becomes a unit grid-cell, the grid-cell list of the next level is created as a child node in order to find clusters of all possible subspaces from the grid-cell. In this way, a maximum n-level grid-cell subspace tree is constructed, and a k-dimensional subspace cluster can be found at the k^th level of the subspace grid-cell tree. Through experiments, it was confirmed that the proposed method uses computing resources more efficiently by expanding only the dense space while maintaining the same accuracy as the existing method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.3836-3854
• /
• 2022

The student group division is benefit for universities to do the student management based on the group profile. With the widespread use of student smart cards on campus, especially where students living in campus residence halls, students' daily activities on campus are recorded with information such as smart card swiping time and location. Therefore, it is feasible to depict the students with the daily activity data and accordingly group students based on objective measuring from their campus behavior with some regular student attributions collected in the management system. However, it is challenge in feature representation due to diverse forms of the student data. To effectively and comprehensively represent students' behaviors for further student group division, we proposed to adopt activity data from student smart cards and student attributes as input data with taking account of activity and attribution relationship types from different perspective. Specially, we propose a novel student group division method based on a multi-view student attribute heterogeneous information network (MSA-HIN). The network nodes in our proposed MSA-HIN represent students with their multi-dimensional attribute information. Meanwhile, the edges are constructed to characterize student different relationships, such as co-major, co-occurrence, and co-borrowing books. Based on the MSA-HIN, embedded representations of students are learned and a deep graph cluster algorithm is applied to divide students into groups. Comparative experiments have been done on a real-life campus dataset collected from a university. The experimental results demonstrate that our method can effectively reveal the variability of student attributes and relationships and accordingly achieves the best clustering results for group division.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.2
• /
• pp.561-568
• /
• 2023

Various defects occur in concrete structures due to internal and external environments. If there is a defect, it is important to efficiently identify and maintain it because there is a problem with the structural safety of concrete. However, recent deep learning research has focused on cracks in concrete, and studies on exfoliation and contamination are lacking. In this study, focusing on exfoliation and contamination, which are difficult to label, four models were developed and their performance evaluated through unlabelling method, filtering method, the convergence of transfer learning based k-means clustering. As a result of the analysis, the convergence model classified the defects in the most detail and could increase the efficiency compared to direct labeling. It is hoped that the results of this study will contribute to the development of deep learning models for various types of defects that are difficult to label in the future.

• International Journal of Advanced Culture Technology
• /
• v.11 no.4
• /
• pp.223-232
• /
• 2023

The purpose of this study is to determine the impact of social network sports community consciousness on loyalty and sports attitude. In order to achieve this research purpose, the population of the study was selected as adult men and women over the age of 20 who are active in the social network sports community in 2022. The sampling method used cluster random sampling to select a total of 300 people, 150 men and 150 women, as research subjects. The survey tool used was the questionnaire method, and the questionnaire whose reliability and validity had been verified in previous studies at home and abroad was used by requoting, modifying, or supplementing it to suit the purpose of this study. It was also structured on a 5-point scale. Frequency analysis, factor analysis, reliability analysis, simple regression analysis, and multiple regression analysis were performed on the collected data using the statistical program SPSS Windows 20.0 Version. The results obtained through this process are as follows. First, social network sports community consciousness was found to have a partial effect on loyalty. Second, social network sports community consciousness was found to have a partial effect on sports attitudes. Third, social network sports community loyalty was found to have a partial effect on sports attitudes. Considering these results, various activities such as decision-making process, relationship formation, and opinion expression of modern people are carried out by the O-line community. In addition, while in the past it was a format that led from offline activities to online activities, currently, there are more and more formats that lead from online activities to offline activities. Therefore, modern people's SNS sports community activities provide many experiences, which creates a sense of community and sports attitudes are formed based on this. This can be said to lead to loyal activities.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.3
• /
• pp.129-136
• /
• 2006

For the environmental noise assessments. A weighted equivalent noise level (LeqA) is used to measure the time varying environmental noise. However, it is not appropriately reflect various environmental noise features and human emotions. The human perception of the noise is affected largely by the psychoacoustic characteristics of noise as well as the sound pressure level In this study, the effective factors of noise qualify are analyzed using the subjective assessment and statistical analysis of environmental noise, such as road traffic noise. construction site noise, noise in daily living. and other. The analysis methodology is composed to three steps as follows : firstly, the values of the sound qualify metrics of various noise sources were analyzed. And to classify the noise sources, we conducted a cluster analysis using sound quality metrics. Secondly, subjective jury testing was carried out using the methods of paired comparisons and semantic differential. Finally, the correlation between the subjective parameters and the noise quality metrics were analyzed. As a result. the human perception characteristics of the various environmental noise are described in some physical parameters of the noise qualify metrics.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.34 no.4
• /
• pp.1105-1116
• /
• 2014

A stochastic rainfall model, NSRPM (Neyman-Scott Rectangular Pulse Model), is able to reflect the cluster characteristics of rainfall events which is unable in the RPM (Rectangular Pulse Model). Therefore NSRPM has advantage in the hydrological applications. The NSRPM consists of five model parameters and the parameters are estimated using optimization techniques such as DFP (Davidon-Fletcher-Powell) method and genetic algorithm. However the DFP method is very sensitive in initial values and is easily converge to local minimum. Also genetic algorithm has disadvantage of long computation time. Nelder-Mead method has several advantages of short computation time and no need of a proper initial value. In this study, the applicability of parameter estimation methods was evaluated using rainfall data of 59 national rainfall networks from 1973-2011. Overall results demonstrated that accuracy in parameter estimation is in the order of Nelder-Mead method, genetic algorithm, and DFP method.