• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.905-922
• /
• 2021

Cluster analysis is one of unsupervised learning techniques used for discovering clusters when there is no prior knowledge of group membership. K-means, one of the commonly used cluster analysis techniques, may fail when the number of variables becomes large. In such high-dimensional cases, it is common to perform tandem analysis, K-means cluster analysis after reducing the number of variables using dimension reduction methods. However, there is no guarantee that the reduced dimension reveals the cluster structure properly. Principal component analysis may mask the structure of clusters, especially when there are large variances for variables that are not related to cluster structure. To overcome this, techniques that perform dimension reduction and cluster analysis simultaneously have been suggested. This study proposes probabilistic reduced K-means, the transition of reduced K-means (De Soete and Caroll, 1994) into a probabilistic framework. Simulation shows that the proposed method performs better than tandem clustering or clustering without any dimension reduction. When the number of the variables is larger than the number of samples in each cluster, probabilistic reduced K-means show better formation of clusters than non-probabilistic reduced K-means. In the application to a real data set, it revealed similar or better cluster structure compared to other methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.33 no.8C
• /
• pp.633-641
• /
• 2008

Since WSNs encounter attacks, such as jamming or eavesdropping without physical access occurs, security is one of the important requirements for WSNs. The key pre-distribution scheme that was recently researched for advance of security in WSNs distributes the keys and probability with the use of q-composite random key pre-distribution method, but there is a high probability that no key shared between sensor nodes, and it takes a lot of time and energy to find out the shared key. Therefore, it is not suitable for WSNs. In order to enhance stability of a node that plays a role of gateway without depending on probabilistic key, this paper proposesa key pre-distribution scheme combined with random key pre-distribution scheme and double hash chain. Since the proposed scheme can maintain a small storage place and strong security strengths, it is more efficient than the existing schemes with the same security strengths. In addition, since it uses a small size of key generation key set, it can reduce a great deal of storage overhead.

• Management & Information Systems Review
• /
• v.34 no.5
• /
• pp.71-92
• /
• 2015

This study analyzes the impact of Corporate Social Responsibility (CSR) activities on financial performances using Bayesian Network. The research tries to overcome the issues of the uniform assumption of a linear function between financial performance and CSR activities in multiple regression analysis widely used in previous studies. It is required to infer a causal relationship between activities of CSR which have an impact on the financial performances. Identifying the relationship would empower the firms to improve their financial performance by informing the decision makers about the different CSR activities that influence the financial performance of the firms. This research proposes General Bayesian Network (GBN) and presents Markov Blanket induced from GBN. It is empirically demonstrated that all the proposals presented in this study are statistically significant by the results of the research conducted by Korean Economic Justice Institute (KEJI) under Citizen's Coalition for Economic Justice (CCEJ) which investigated approximately 200 companies in Korea based on Korean Economic Justice Institute Index (KEJI index) from 2005 to 2011. The Bayesian Network to effectively infer the properties affecting financial performances through the probabilistic causal relationship. Moreover, I found that there is a causal relationship among CSR activities variable; that is Environment protection is related to Customer protection, Employee satisfaction, and firm size; Soundness is related to Total CSR Evaluation Score, Debt-Assets Ratio. Though the what-if analysis, I suggest to the sensitive factor among the explanatory variables.

• Resources Recycling
• /
• v.27 no.1
• /
• pp.84-91
• /
• 2018

In this study, a novel soft information based most probable classification scheme is proposed for sorting recyclable metal alloys with laser induced breakdown spectroscopy (LIBS). Regression analysis with LIBS captured spectrums for estimating concentrations of common elements can be efficient for classifying unknown arbitrary metal alloys, even when that particular alloy is not included for training. Therefore, partial least square regression (PLSR) is employed in the proposed scheme, where spectrums of the certified reference materials (CRMs) are used for training. With the PLSR model, the concentrations of the test spectrum are estimated independently and are compared to those of CRMs for finding out the most probable class. Then, joint soft information can be obtained by assuming multi-variate normal (MVN) distribution, which enables to account the probability measure or a prior information and improves classification performance. For evaluating the proposed schemes, MVN soft information is evaluated based on PLSR of LIBS captured spectrums of 9 metal CRMs, and tested for classifying unknown metal alloys. Furthermore, the likelihood is evaluated with the radar chart to effectively visualize and search the most probable class among the candidates. By the leave-one-out cross validation tests, the proposed scheme is not only showing improved classification accuracies but also helpful for adaptive post-processing to correct the mis-classifications.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.23 no.1
• /
• pp.9-16
• /
• 2017

As a preliminary study of enter or leaving traffic patterns of the Korea main port, port Management Information System (Port-MIS) data was used to check the volume of vessels entering and leaving the port of Busan, and three consecutive days from each seasons were selected for study. Selected 12-day General Information Center on Maritime Safety & Security (GICOMS) data was also used to analyze the traffic pattern in the main traffic lane of Busan port for dangerous goods carrier. Also, the distance between dangerous goods carriers and Oryukdo breakwater of east breakwater in the main traffic lane was analyzed. Collision probability was estimated using the cumulative probability distribution function of the normal distribution for the maritime traffic safety audit scheme based on the assumption that a ship's trajectory has a normal distribution for a section of the route. However, in case of entry or leaving thorough the Oryukdo breakwater and entry thorough the east breakwater, ship's sailing trajectories were revealed not to follow a normal distribution via regularity testing using a KS-test and SW-test. Especially in the north port, the tendency of the right side of the ship to pass was remarkable. It is desirable to develop a traffic model suitable for the characteristics of the port rather than to apply general traffic theories, and to apply this model to a maritime traffic safety diagnosis, so further research is needed.

• Journal of the Korean Society of Radiology
• /
• v.13 no.3
• /
• pp.407-414
• /
• 2019

Recently, after the reorganization as the basis of NCS education, various learning methods are being sought for improving the basic occupational ability and job ability required by NCS, and the evaluation method accordingly is urgently needed. The purpose of this study was to evaluate the applicability of meta-cognitive learning and Havruta learning as evaluation cases in order to improve the job skills and basic skills required in the NCS curriculum. As a result, the meta-cognitive learning response sample statistic showed an average of 2.6883 when the pre-meta-cognitive learning questionnaire was a 5-point scale, and an average of 4.2468 after the meta-cognitive learning questionnaire. The correlation coefficient was 0.782 and the significance probability was 0.045. In the case of the Havruta learning correspondence sample statistic, the average of 3.1515 when the preliminary Havruta learning questionnaire was a 5 point scale and the average of the post-Havruta learning questionnaire was 4.3853, which was improved by 1.23 points. The correlation coefficient was 0.631 and the significance probability was 0.049. Meta-cognitive learning and Havruta learning were found to be correlated. The mean of meta cognition was 3.4675 and the mean of Havruta was 3.7684. Metacognitive learning and Havruta learning were -0.042 And there was no statistically significant difference. Therefore, the learning method to improve the job ability should be applied considering the characteristics of the subject.

• Journal of KIISE:Information Networking
• /
• v.33 no.3
• /
• pp.193-200
• /
• 2006

The security feature is essential in wireless sensor network such as intrusion detection or obstacle observation. Sensor nodes must have shared secret between nodes to support security such as privacy. Many methods which provide key pre-distribution need too many keys or support poor security. To solve this problem, probabilistic key pre-distribution is proposed. This method needs a few keys and use probabilistic method to share keys. However, this method does not guarantee key sharing between nodes, and neighbor nodes nay not communicate each other. It leads to waste of network resource such as inefficient routing, extra routing protocol. In this paper, we propose new key distribution method using quorum system which needs a few keys and guarantee key sharing between nodes. We also propose extension of the method which needs fewer keys and guarantee key sharing when node deployment knowledge is well known.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.549-558
• /
• 2020

In this paper, the method is discussed such that the robustness of automatic initiated narrowband target tracking is improved in passive sonar. In the case of automatic tracking initiation as target in passive sonar, due to a number of clutter, the clutter is initiated as target and tracked which prohibits the operation capability. The associated probability and information entropy of measurements, extracted from detection data, is calculated to keep going on automatic target initiation and tracking of true target, but reduce the automatic initiation and tracking of clutter. If the association probability and information entropy of the extracted measurements is satisfied for the predefined conditions, the procedure of automatic initiation begins. Using sea-trial data, simulations are executed and the results from the proposed method indicate that it keeps the automatic target initiation and tracking of true target and suppresses the automatic target initiation and tracking of clutters in contrary to the conventional method.

• Journal of the Korea Institute of Building Construction
• /
• v.17 no.6
• /
• pp.535-544
• /
• 2017

This paper analyzes the relations between design errors, prevented by building information modeling (BIM)-based design validation, and their impacts in order to identify critical consideration factors for implementing BIM-based design validation in architecture, engineering, and construction (AEC) projects. More than 800 design errors detected by BIM-based design validation in two BIM-based projects in South Korea are categorized according to their causes (illogical error, discrepancy, and missing item) and work types (structure, architecture, and mechanical, electrical, and plumbing (MEP)). The probabilistic relations among the independent variables, including the causes and work types of design errors, and the dependent variables, including the project delays, cost overruns, low quality, and rework generation that can be caused by these errors, are analyzed using logistic regression. The characteristics of each design error are analyzed by means of face-to-face interviews with practitioners. According to the results, the impacts of design error causes in predicting the probability values of project delays, cost overruns, low quality, and rework generation were statistically meaningful.

• Journal of the Korean Geographical Society
• /
• v.51 no.4
• /
• pp.525-539
• /
• 2016

This study attempts to predict the changes in the spatial distribution of occupational employment in Korea by applying Markov Chain Model. For the purpose we analyze the job-related migration pattern and estimate the transition probability with the last six years job-related migration data. By applying the Chapman-Kolmogorov equation based on the transition probability, we predict the changes in the spatial distribution of occupational employment for the next ten years. The result reveals that the employment of professional jobs is predicted to increase at every city and region except Seoul, while the employment of elementary labor jobs is predicted to increase slightly in Seoul. In particular, Gangwon-do and Chuncheongdo are predicted to increase in the employment of all occupational jobs.