• Journal of Information Processing Systems
• /
• 제13권2호
• /
• pp.400-416
• /
• 2017

Music emotion is an important component in the field of music information retrieval and computational musicology. This paper proposes an approach for automatic emotion classification, based on rough set (RS) theory. In the proposed approach, four different sets of music features are extracted, representing dynamics, rhythm, spectral, and harmony. From the features, five different statistical parameters are considered as attributes, including up to the $4^{th}$ order central moments of each feature, and covariance components of mutual ones. The large number of attributes is controlled by RS-based approach, in which superfluous features are removed, to obtain indispensable ones. In addition, RS-based approach makes it possible to visualize which attributes play a significant role in the generated rules, and also determine the strength of each rule for classification. The experiments have been performed to find out which audio features and which of the different statistical parameters derived from them are important for emotion classification. Also, the resulting indispensable attributes and the usefulness of covariance components have been discussed. The overall classification accuracy with all statistical parameters has recorded comparatively better than currently existing methods on a pair of datasets.

• 대한산업공학회지
• /
• 제20권1호
• /
• pp.71-85
• /
• 1994

Overhead cranes which carry heavy items in construction or production areas consist of structure and electric modules. More than 80% of breakdowns bring out of the electric module. As operators do not know all about cranes, it sometimes takes much time to repair the cranes. In order to resolve this problem, the expert system which can diagnose causes of faults and give instructions for repair to operators, has been developed. The scope of the paper is limited to the electric module. First of all, analyzing symptoms and causes, we have developed a rule base with the expert system shell, EXSYS. Furthermore, for the facility maintenance including repair instructions against the causes, the instruction data base was developed with FOXPRO. On the other hand, for the help of user's understanding the fault causes, the graphic animation module which shows malfunctioning component ports or motions in 3D was developed with the graphic software, TOPAS VGA.

• 산업공학
• /
• 제12권3호
• /
• pp.468-479
• /
• 1999

Lean production system can be defined as customer(product)-oriented production system with small lot size and flow-shop layout based on the JIT(Just-in-time) principles. In this paper, we introduce a case example of implementation of the Lean product ion system for manufacturing line of electronic component which has both machine processes and manual jobs. We also investigate the issues of implementing the Lean production system with the viewpoints of layout design scheme and JIT management rules. In the layout design, we propose the cell-line which has flow-shop layout with small lot size. In the management rules, the superior cell rule is applied in order to boost the needs of kaisen up. As the results of implementing the Lean production system, production lead time is decreased from 5 days to 1.5 days and also productivity and quality level arc surprisingly increased.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.109.6-109
• /
• 2002

In this paper, the scheme of fuzzy system construction using GA(genetic algorithm) and FCM(Fuzzy c-means) clustering algorithm is proposed for TSK(Takagi-Sugeno-Kang) type fuzzy system. in the structure identification, input data is trans-formed by PCA(Principal Component Analysis) to reduce the correlation among input data components. And then, the number of fuzzy rule is obtained by a given performance criterion. In the parameter identification, the premise parameters are optimally searched by GA. On the other hand, the consequent parameters are estimated by RLSE(Recursive Least Square Estimate) to reduce the search space. From this, one can systematically obtain optimal parameter and the v..

• 한국컴퓨터정보학회논문지
• /
• 제23권10호
• /
• pp.173-180
• /
• 2018

In this paper, we analyzed key keywords and research themes in the field of defense research using keyword network analysis and tried to grasp the whole knowledge structure. To do this, we extracted data from 2,165 research data from defense related research institutes from 2010 to 2017 and applied the Pareto rule to the number of abstracts of words and the number of links between words, We extracted a total of 2,303 words based on the criterion and extracted 204 final key words through component analysis. By analyzing the centrality and cohesiveness through these key words, we confirmed the concept of core research in the defense field and derived a total of 7 large groups and 16 small groups of each group in the knowledge structure of the defense area.

• Computers and Concrete
• /
• 제18권2호
• /
• pp.179-199
• /
• 2016

The aim of this study is to investigate arrival direction effects of travelling waves on non-linear seismic response of arch dams. It is evident that the seismic waves may reach on the dam site from any direction. Therefore, this study considers the seismic waves arrive to the dam site with different angles, ${\theta}=0^{\circ}$, $15^{\circ}$, $30^{\circ}$, $45^{\circ}$, $60^{\circ}$, $75^{\circ}$, and $90^{\circ}$ for non-linear analysis of arch dam-water-foundation interaction system. The N-S, E-W and vertical component of the Erzincan earthquake, on March 13, 1992, is used as the ground motion. Dam-water-foundation interaction is defined by Lagrangian approach in which a step-by-step integration technique is employed. The stress-strain behavior of the dam concrete is idealized using three-dimensional Drucker-Prager model based on associated flow rule assumption. The program NONSAP is employed in response calculations. The time-history of crest displacements and stresses of the dam are presented. The results obtained from non-linear analyses are compared with that of linear analyses.

• Clinics in Shoulder and Elbow
• /
• 제25권3호
• /
• pp.244-254
• /
• 2022

Radiographic osteolysis after total shoulder arthroplasty (TSA) remains a challenging clinical entity, as it may not initially manifest clinically apparent symptoms but can lead to clinically important complications, such as aseptic loosening. A thorough consideration of medical history and physical examination is essential to rule out other causes of symptomatic TSA-namely, periprosthetic joint infection-as symptoms often progress to vague pain or discomfort due to subtle component loosening. Once confirmed, nonoperative treatment of osteolysis should first be pursued given the potential to avoid surgery-associated risks. If needed, the current surgical options include glenoid polyethylene revision and conversion to reverse shoulder arthroplasty. The current article provides a comprehensive review of the evaluation and management of osteolysis after TSA through an evidence-based discussion of current concepts.

• 한국컴퓨터산업학회논문지
• /
• 제2권9호
• /
• pp.1175-1182
• /
• 2001

원투원 마케팅(데이터베이스 마케팅 또는 관계 마케팅)은 컴퓨터의 발전과 더불어 기업 및 고객에게 이익을 가져올 것이며, 또한 고객의 세일 및 광고에 변화를 가져올 여러 분야 중의 하나이다. 인터넷 쇼핑몰에서 지능적인 고객 서비스의 일환으로, 본 논문에서는 데이터 마이닝 기법으로 잘 알려진 장바구니 분석을 이용한 개인화 된 광고를 제공하는 기법을 제시하고자 한다. 추천 기법의 핵심적인 이론으로 통계학, 데이터 마이닝, 인공 지능, 규칙 기반 매칭 등이 있다. 개인화 된 추천을 위한 규칙 기반 관점에서, 개인화를 위한 마케팅 규칙은 일반적으로 마케팅 전문가로부터 추출되어 고객의 데이터를 갖고 추정한다. 그러나 마케팅 전문가로부터 규칙을 추출하기란 매우 어려울 뿐만 아니라, 작성된 지식 기반 규칙을 검증하고 유지하기도 어렵다. 본 논문에서는 장바구니 분석 기법을 이용하여, 크로스 세일 마케팅 규칙을 추출한 뒤, 고객이 인터넷 쇼핑몰에 방문했을 때 개인화 된 광고를 제공하는데 초점을 두기로 한다.

• 한국산학기술학회논문지
• /
• 제13권2호
• /
• pp.872-877
• /
• 2012

본 논문은 한글 모음의 구조적 특징을 이용하여 자연영상에 포함된 한글 문자영역을 검출하는 기법을 제안하였다. 자연 영상을 명도영상으로 변환하고 에지 및 연결요소 기반 방법으로 특징값을 추출하며, 추출된 특징값은 필터링을 수행하여 한글 문자의 특징에 맞지 않는 특징값을 제거하여 한글 문자영역 병합을 위한 후보를 선정한다. 선정된 후보 특징값은 한글 자소 병합 알고리즘으로 하나의 문자로 병합하여 후보 문자영역으로 검출하고, 한글 문자 유형 판별 알고리즘으로 한글 문자영역 여부를 판별함으로서 최종적인 한글 문자영역을 검출한다. 실험결과, 복잡한 배경을 갖고 다양한 환경에서 촬영된 영상에서 한글 문자영역을 효과적으로 검출하였고, 제안한 문자영역 검출 방법은 향상된 검출 결과를 보여 주었다.

• 대한의용생체공학회:의공학회지
• /
• 제31권6호
• /
• pp.456-463
• /
• 2010

In the present study, we proposed a new subspace scanning algorithm to enhance the spatial resolution of electroencephalography (EEG) and magnetoencephalography(MEG) source localization. Subspace scanning algorithms, represented by the multiple signal classification (MUSIC) algorithm and the first principal vector (FINE) algorithm, have been widely used to localize asynchronous multiple dipolar sources in human cerebral cortex. The conventional MUSIC algorithm used principal component analysis (PCA) to extract the noise vector subspace, thereby having difficulty in discriminating two or more closely-spaced cortical sources. The FINE algorithm addressed the problem by using only a part of the noise vector subspace, but there was no golden rule to determine the number of noise vectors. In the present work, we estimated a non-orthogonal signal vector set using independent component analysis (ICA) instead of using PCA and performed the source scanning process in the signal vector subspace, not in the noise vector subspace. Realistic 2D and 3D computer simulations, which compared the spatial resolutions of various algorithms under different noise levels, showed that the proposed ICA-MUSIC algorithm has the highest spatial resolution, suggesting that it can be a useful tool for practical EEG/MEG source localization.