• International Journal of Contents
• /
• 제14권4호
• /
• pp.24-29
• /
• 2018

EEG-based brain-computer interfaces has focused on explicitly expressed intentions to assist physically impaired patients. For EEG-based-computer interfaces to function effectively, it should be able to understand users' implicit information. Since it is hard to gather EEG signals of human brains, we do not have enough training data which are essential for proper classification performance of implicit intention. In this paper, we improve the subject independent classification of implicit intention through the generation of additional training data. In the first stage, we perform the PCA (principal component analysis) of training data in a bid to remove redundant components in the components within the input data. After the dimension reduction by PCA, we train ICA (independent component analysis) network whose outputs are statistically independent. We can get additional training data by adding Gaussian noises to ICA outputs and projecting them to input data domain. Through simulations with EEG data provided by CNSL, KAIST, we improve the classification performance from 65.05% to 66.69% with Gamma components. The proposed sample generation method can be applied to any machine learning problem with fewer samples.

• International Journal of Contents
• /
• 제12권3호
• /
• pp.12-16
• /
• 2016

Brain computer interfaces (BCI) usually have focused on classifying the explicitly-expressed intentions of humans. In contrast, implicit intentions should be considered to develop more intelligent systems. However, classifying implicit intention is more difficult than explicit intentions, and the difficulty severely increases for subject independent classification. In this paper, we address the subject independent classification of implicit intention based on electroencephalography (EEG) signals. Among many machine learning models, we use the support vector machine (SVM) with radial basis kernel functions to classify the EEG signals. The Fisher scores are evaluated after extracting the gamma, beta, alpha and theta band powers of the EEG signals from thirty electrodes. Since a more discriminant feature has a larger Fisher score value, the band powers of the EEG signals are presented to SVM based on the Fisher score. By training the SVM with 1-out of-9 validation, the best classification accuracy is approximately 65% with gamma and theta components.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제53권12호
• /
• pp.834-838
• /
• 2004

This paper proposes an algorithm for EEG pattern classification using the Multi-template method, which is a kind of speaker adaptation method for speech signal processing. 10-channel EEG signals are collected in various environments. The linear prediction coefficients of the EEGs are extracted as the feature parameter of human sensibility. The human sensibility classification algorithm is developed using neural networks. Using EEGs of comfortable or uncomfortable seats, the proposed algorithm showed about 75% of classification performance in subject-independent test. In the tests using EEG signals according to room temperature and humidity variations, the proposed algorithm showed good performance in tracking of pleasantness changes and the subject-independent tests produced similar performances with subject-dependent ones.

• Journal of Computing Science and Engineering
• /
• 제7권2호
• /
• pp.132-138
• /
• 2013

Multi-modal techniques have received increasing interest in the neuroscientific and brain computer interface (BCI) communities in recent times. Two aspects of multi-modal imaging for BCI will be reviewed. First, the use of recordings of multiple subjects to help find subject-independent BCI classifiers is considered. Then, multi-modal neuroimaging methods involving combined electroencephalogram and near-infrared spectroscopy measurements are discussed, which can help achieve enhanced and robust BCI performance.

• 한국기록관리학회지
• /
• 제18권4호
• /
• pp.99-124
• /
• 2018

본 연구는 국가보훈처의 '공훈전자사료관' 컬렉션을 기반으로 독립 운동 컬렉션에 대한 출처별, 주제별, 자료유형별 분류 체계를 제안하였다. 먼저 출처별 분류 체계는 계층별로 구성하였으며 출처별로 분류된 기록물들은 관련 주제어에 따라 2차 분류한 후, 17가지 자료유형별로 분류하였고 '출처-주제-자료유형'의 인용 순서로 기술하였다. 또한 주제어를 바탕으로 한 귀납적 방법을 이용하여 유의미한 컬렉션을 도출하였다. 이를 통해 공훈전자사료관 컬렉션을 출처별, 주제별, 자료유형별로 이용자가 기록물을 쉽게 찾을 수 있도록 분류하였다. 본 연구의 결과는 독립 운동 DB 자료의 기록물 검색을 지원하기 위한 분류 체계로서 이용자의 접근성을 높이고 독립 운동 기록물 서비스를 확장하는데 기초가 될 것이다.

• 한국비블리아학회지
• /
• 제27권3호
• /
• pp.25-50
• /
• 2016

기록관리학이 독자적인 학문영역으로 발전해가고 있다. 하지만 KDC를 비롯한 기존의 문헌분류표에서는 분류 항목 배열이 기록관리학의 특성을 제대로 반영하지 못하여 분류항목의 전개가 불합리한 부분이 있으며, 주제영역별 세목전개의 재배치가 필요하다. 따라서 이 연구에서는 기록관리학의 학문 특성에 따라 주제영역을 기록관리 일반, 기록관리 법 정책, 기록물의 수집 선별 평가, 기록물의 조직, 기록정보서비스, 기록물 관리 및 보존, 기록관 운영, 기록관리 기관 등 8개 영역으로 설정하고, 현대 주요 문헌분류표인 KDC, DDC, NDC, UDC, LCC의 분류체계를 분석한 후, "대한민국 국가서지"의 기록관리학 분야 유별 자료현황 및 주제어 분석 결과를 반영하여 KDC 기록관리학 분야의 분류체계 수정 전개 방안을 제시하였다. 기록관리학 관련 주제영역 8개 분야의 내용은 KDC 028로 통합할 수 있도록 변경하였다.

• 한국도서관정보학회지
• /
• 제14권
• /
• pp.71-98
• /
• 1987

After 1970s, because of an increasing peoples for higher education and graduate school-oriented education system, quantities of thesis were produced for short term and they began to be an important part of gift materials in university libraries. As an unpublished documents, thesis is narrow in subject, deep in content, irregular is page, incomplete in binding and produce limited-edition during short time simultaneously at most institutions. So most libraries are in difficulties for acquisition processing and circulation. Because of an increasing number of thesis will be produce and cutback of budget, shortage of staff, library service for user will be difficult and rational and efficient management is absolutely essential to library. In form and content, thesis is in distinction with other library materials, they must be handler as an independent item and library will seek an a n.0, ppropriate processing and using method of them. Analysis and synthesis of this study are summarized as follows. 1. In acquisition of thesis, it is desirable that they has an independent accession book with a simplified processing procedure and the binding of them is desirable together with subject field(major or department) by institutions. 2. In classification and cataloging of thesis, it is rational that library use the same classification scheme as other materials and expand in details. Simplified catalog will be reduce the time and/or personnel problem than using the traditional KCR or AACR. 3. As an retrieval tool, author, title, shelf and subject catalog must be prepared in thesis room. Index of thesis will be available for retrieval with the trend of union list and Korean periodicals index (National Assembly Library, Republic of Korea) must include the thesis. 4. Because of the nature and characteristics of thesis, library has to equip an independent room and open stack for the a n.0, ppropriate retrieval and frequent use. Qualified librarian must serve for the efficient circulation service.

• International journal of advanced smart convergence
• /
• 제9권1호
• /
• pp.193-201
• /
• 2020

The main objective of this study is to investigate the impact of additional modalities on the performance of emotion recognition using speech, facial expression and physiological measurements. In order to compare different approaches, we designed a feature-based recognition system as a benchmark which carries out linear supervised classification followed by the leave-one-out cross-validation. For the classification of four emotions, it turned out that bimodal fusion in our experiment improves recognition accuracy of unimodal approach, while the performance of trimodal fusion varies strongly depending on the individual. Furthermore, we experienced extremely high disparity between single class recognition rates, while we could not observe a best performing single modality in our experiment. Based on these observations, we developed a novel fusion method, called parametric decision fusion (PDF), which lies in building emotion-specific classifiers and exploits advantage of a parametrized decision process. By using the PDF scheme we achieved 16% improvement in accuracy of subject-dependent recognition and 10% for subject-independent recognition compared to the best unimodal results.

• Journal of applied mathematics & informatics
• /
• 제8권1호
• /
• pp.9-26
• /
• 2001

In some previous papers the author extended two algorithms proposed by Z. Kovarik for approximate orthogonalization of a finite set of linearly independent vectors from a Hibert space, to the case when the vectors are rows (not necessary linearly independent) of an arbitrary rectangular matrix. In this paper we describe combinations between these two methods and the classical Kaczmarz’s iteration. We prove that, in the case of a consistent least-squares problem, the new algorithms so obtained converge ti any of its solutions (depending on the initial approximation). The numerical experiments described in the last section of the paper on a problem obtained after the discretization of a first kind integral equation ilustrate the fast convergence of the new algorithms. AMS Mathematics Subject Classification : 65F10, 65F20.

• 대한치과교정학회지
• /
• 제21권1호
• /
• pp.113-130
• /
• 1991

This study was undertaken to investigate soft tissue profile changes by orthodontic treatment in female patients. Traditional cephalometric appraisal yields data of dubious scientific value, the soft tissue profile forms were evaluated by finite element method. The subject was divided into three groups according to Angle's classification and each group was composed of 25 female patients averaged aged 12-14 years at the start of treatment. The changes in soft tissue form were evaluated by computing the degree of distortion in each triangle after treatment compared with the triangle before treatment. The conclusions were as follows; 1. The soft tissue profile forms were evaluated by finite element method and independent evaluation of each element by local changes was possible. 2. Maximum and minimum principal strains showed marked variability depending on the particular finite element and each group and Class II, III sample was greater than Class I sample. 3. Soft tissue size changes as a result of orthodontic treatment was not related to those of shape. 4. Soft tissue changes by orthodontic treatment were variable in individual patient, and were not related to Angle's classification.