• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2003.11a
• /
• pp.729-732
• /
• 2003

In this paper, the transformed logo image is used as a watermark image, then the extraction rate is good when the logo image is transformed into randomly distribute Image domain. The robustness is good resist of clipping and noise.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.8
• /
• pp.1373-1379
• /
• 2006

Moving object extraction is one of key research topics for various video services. In this study, a new moving object extraction algorithm is introduced to extract objects using block motion vectors in video data. To do this, 1) a maximum a posteriori probability and Gibbs random field are used to obtain real block motion vectors,2) a 2-D histogram technique is used to determine a global motion, 3) additionally, a block segmentation is fellowed. In the computer simulation results, the proposed technique shows a good performance.

• Proceedings of the Korean Institute of IIIuminating and Electrical Installation Engineers Conference
• /
• 2007.05a
• /
• pp.366-368
• /
• 2007

In this paper, the Time-varying Failure Rate (TFR) of power distribution system equipment is extracted from the recorded failure data of Korea Electric Power Corporation (KEPCO). For TFR extraction, it is used that the fault data accumulated by KEPCO during 10 years. The TFR is approximated to bathtub curve using the exponential (random failure) and Weibull (aging failure) distribution function. In addition, Kaplan-Meier estimation is applied to TFR extraction because of incomplete failure data of KEPCO. Finally, Probability plot and regression analysis is applied. It is presented that the extracted TFR is more effective and useful than Mean Failure Rate (MFR) through the comparison between TFR and MFR.

• Proceedings of the Korean Information Science Society Conference
• /
• 2011.06c
• /
• pp.258-261
• /
• 2011

Living in the age of information people face problems related to information overload. Information is easy to produce, store and distribute through various communication channels, one of which is emails. With the appearance of the mobile devices, such as smart phones and tabs, people can have access to email inbox at any moment of time from everywhere. In this paper we present information extraction system with a specific goal of extracting meeting location from the announcement of seminar or conference. We apply a machine learning method (conditional random fields, CRF), train the system using annotated corpus of seminar and conference announcements and validate results by applying various extracted correction rules and patterns. Furthermore, we normalize extracted location, and reference using geo-coding databases, OpenStreetMap and Wikipedia resources to determine real geographical coordinates.

• Journal of the Korean Society for information Management
• /
• v.40 no.3
• /
• pp.99-118
• /
• 2023

The objective of this study is to identify performance factors in the automatic classification of records by utilizing metadata that contains the contextual information of records. For this study, we collected 97,064 records of original textual information from Korean central administrative agencies in 2022. Various classification algorithms, data selection methods, and feature extraction techniques are applied and compared with the intent to discern the optimal performance-inducing technique. The study results demonstrated that among classification algorithms, Random Forest displayed higher performance, and among feature extraction techniques, the TF method proved to be the most effective. The minimum data quantity of unit tasks had a minimal influence on performance, and the addition of features positively affected performance, while their removal had a discernible negative impact.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.205-213
• /
• 2024

This study uses a large language model (LLM) to identify Aristotle's rhetorical principles (ethos, pathos, and logos) in COVID-19 information on Naver Knowledge-iN, South Korea's leading question-and-answer community. The research analyzed the differences of these rhetorical elements in the most upvoted answers with random answers. A total of 193 answer pairs were randomly selected, with 135 pairs for training and 58 for testing. These answers were then coded in line with the rhetorical principles to refine GPT 3.5-based models. The models achieved F1 scores of .88 (ethos), .81 (pathos), and .69 (logos). Subsequent analysis of 128 new answer pairs revealed that logos, particularly factual information and logical reasoning, was more frequently used in the most upvoted answers than the random answers, whereas there were no differences in ethos and pathos between the answer groups. The results suggest that health information consumers value information including logos while ethos and pathos were not associated with consumers' preference for health information. By utilizing an LLM for the analysis of persuasive content, which has been typically conducted manually with much labor and time, this study not only demonstrates the feasibility of using an LLM for latent content but also contributes to expanding the horizon in the field of AI text extraction.

• Journal of Information Processing Systems
• /
• v.14 no.5
• /
• pp.1167-1175
• /
• 2018

Microarray data plays an essential role in diagnosing and detecting cancer. Microarray analysis allows the examination of levels of gene expression in specific cell samples, where thousands of genes can be analyzed simultaneously. However, microarray data have very little sample data and high data dimensionality. Therefore, to classify microarray data, a dimensional reduction process is required. Dimensional reduction can eliminate redundancy of data; thus, features used in classification are features that only have a high correlation with their class. There are two types of dimensional reduction, namely feature selection and feature extraction. In this paper, we used k-means algorithm as the clustering approach for feature selection. The proposed approach can be used to categorize features that have the same characteristics in one cluster, so that redundancy in microarray data is removed. The result of clustering is ranked using the Relief algorithm such that the best scoring element for each cluster is obtained. All best elements of each cluster are selected and used as features in the classification process. Next, the Random Forest algorithm is used. Based on the simulation, the accuracy of the proposed approach for each dataset, namely Colon, Lung Cancer, and Prostate Tumor, achieved 85.87%, 98.9%, and 89% accuracy, respectively. The accuracy of the proposed approach is therefore higher than the approach using Random Forest without clustering.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.7 no.6
• /
• pp.979-994
• /
• 2015

The mode shapes of a segmented hull model towed in a model basin were predicted using both the Proper Orthogonal Decomposition (POD) and cross random decrement technique. The proper orthogonal decomposition, which is also known as Karhunen-Loeve decomposition, is an emerging technology as a useful signal processing technique in structural dynamics. The technique is based on the fact that the eigenvectors of a spatial coherence matrix become the mode shapes of the system under free and randomly excited forced vibration conditions. Taking advantage of the simplicity of POD, efforts have been made to reveal the mode shapes of vibrating flexible hull under random wave excitation. First, the segmented hull model of a 400 K ore carrier with 3 flexible connections was towed in a model basin under different sea states and the time histories of the vertical bending moment at three different locations were measured. The measured response time histories were processed using the proper orthogonal decomposition, eventually to obtain both the first and second vertical vibration modes of the flexible hull. A comparison of the obtained mode shapes with those obtained using the cross random decrement technique showed excellent correspondence between the two results.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.1
• /
• pp.95-102
• /
• 2022

Since the organic light-emitting diodes (OLEDs) have been widely investigated as next-generation displays, it has been successfully commercialized as a flexible and rollable display. However, there is still wide room and demand to improve the device characteristics such as power efficiency and lifetime. To solve this issue, there has been a wide research effort, and among them, the internal and the external light extraction techniques have been attracted in this research field by its fascinating characteristic of material independence. In this study, a micro-nano composite structured external light extraction layer was demonstrated. A reactive ion etching (RIE) process was performed on the surfaces of hexagonally packed hemisphere micro-lens array (MLA) and randomly distributed sphere diffusing films to form micro-nano composite structures. Random nanostructures of different sizes were fabricated by controlling the processing time of the O₂ / CHF₃ plasma. The fabricated device using a micro-nano composite external light extraction layer showed 1.38X improved external quantum efficiency compared to the reference device. The results prove that the external light extraction efficiency is improved by applying the micro-nano composite structure on conventional MLA fabricated through a simple process.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.553-555
• /
• 2001

Markov Random Fields(MRF) 모델은 영상 분할 및 복원 등에 주로 사용되는 확률적 영상모델이다. 본 논문에서는 MRF 모델을 3차원으로 확장하여 분할을 위한 선 필드 모델(Line Field Model)과 움직임 검출을 위한 움직임 필드 모델(Motion Field Model)을 도입하여 동영상 내에서 움직이는 물체의 형상과 움직임을 추정한다. 제안된 방법을 이용하여 한국어 수화 동작에서 손의 형상과 이동방향을 검출하였다. 그 결과 optical flow를 사용하는 방법에 비해서 이동 방향이 왜곡되는 것을 방지하여 보다 정확한 이동 방향을 검출할 수 있었다. 또한 영상 추출의 경우에 있어서도 형상의 윤곽면과 내부가 하나의 라벨(label)로 묶이기 때문에 보다 깨끗한 영상을 추출할 수 있었다.