• 한국정보통신학회논문지
• /
• 제17권3호
• /
• pp.705-711
• /
• 2013

심전도 신호의 QRS 영역은 심장의 질환을 판단하는 중요한 자료로 쓰이는데, 여러 종류의 잡음으로 인해 이를 분석하는데 어려움을 준다. 또한 일반인들의 건강상태를 지속적으로 모니터링 하는 헬스케어 시스템에서는 신호의 실시간 처리가 필요하다. 그리고 생체신호의 특성상 개인 간의 차이가 있음에도 불구하고, 일반적인 ECG 신호의 판단 규칙에 따라 진단을 수행함으로써 성능하락이 나타날 수밖에 없다. 이러한 문제점을 해결하기 위해서는 최소한의 연산량으로 QRS를 검출하고 환자의 특성에 맞게 부정맥을 분류할 수 있는 알고리즘의 설계가 필요하다. 따라서 본 연구에서는 형태연산을 통한 효율적인 QRS 검출과 개인별 정상신호 분류를 위해 해쉬 함수를 적용하여 프로파일링 하였으며, 검출된 QRS 폭과 RR 간격을 이용하여 심실조기수축(PVC)을 분류하는 알고리즘을 개발하였다. 제안한 방법의 우수성을 입증하기 위해 MIT-BIH 부정맥 데이터베이스를 통해 기존 방법과 부정맥 분류 성능을 비교하였다. 성능평가 결과, R파는 평균 99.77%, 정상 신호 분류에 대한 에러율은 0.65%, PVC는 각각 93.29%로 기존 방법에 비해 약 5% 우수하게 나타났다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권12호
• /
• pp.3383-3397
• /
• 2023

Scene graphs serve as semantic abstractions of images and play a crucial role in enhancing visual comprehension and reasoning. However, the performance of Scene Graph Generation is often compromised when working with biased data in real-world situations. While many existing systems focus on a single stage of learning for both feature extraction and classification, some employ Class-Balancing strategies, such as Re-weighting, Data Resampling, and Transfer Learning from head to tail. In this paper, we propose a novel approach that decouples the feature extraction and classification phases of the scene graph generation process. For feature extraction, we leverage a transformer-based architecture and design an adaptive calibration function specifically for predicate classification. This function enables us to dynamically adjust the classification scores for each predicate category. Additionally, we introduce a Distribution Alignment technique that effectively balances the class distribution after the feature extraction phase reaches a stable state, thereby facilitating the retraining of the classification head. Importantly, our Distribution Alignment strategy is model-independent and does not require additional supervision, making it applicable to a wide range of SGG models. Using the scene graph diagnostic toolkit on Visual Genome and several popular models, we achieved significant improvements over the previous state-of-the-art methods with our model. Compared to the TDE model, our model improved mR@100 by 70.5% for PredCls, by 84.0% for SGCls, and by 97.6% for SGDet tasks.

• 기술혁신학회지
• /
• 제10권2호
• /
• pp.183-205
• /
• 2007

본 연구는 기술력평가에 근거해서 중소기업 부실예측 가능성을 사전에 예측할 수 있는 최적 판별 모형을 개발 제안하였다. 판별모형에 포함될 설명변수는 요인분석과 판별모형의 단계별 선택방법에 의하여 선정되었다. 분석결과 선형판별모형이 로지스틱판별모형보다 임계확률 관점에서 적절한 것으로 나타났다. 최적 선형판별모형의 분류 정분류율은 70.4%, 분류 예측력은 67.5%로 나타났다. 최적 선형판별모형의 활용도를 높이기 위해서 확실 범주와 유보범주를 구분할 수 있는 경계값을 설정하였다. 분석결과를 활용하면 기술금융 취급기관은 부실위험 평가와 더불어 기술금융 신청기업의 순위를 부여할 때 유용하게 사용할 수 있을 것으로 기대된다.

• The Journal of Korean Physical Therapy
• /
• 제20권1호
• /
• pp.49-56
• /
• 2008

Purpose: Lack of a valid prognosis of gross motor development in children with cerebral palsy (CP) and the absence of longitudinal data on which to base an opinion in Korea have made it difficult to plan treatment and counsel prognosis issues accurately. The purposes of this study were to examine whether the Gross Motor Function Classification System (GMFCS) is valuable to prognostication about gross motor progress in children with CP in Korea. Methods: Medical records of 61 patients were retrospectively reviewed that visited outpatient department and were diagnosed as CP. Various information was surveyed including CP type, visual acuity, cognitive function, motor acquisition age, ambulatory status, development curves of Gross Motor Function Measure (GMFM) according to each of the 5 level of GMFCS. All of them were compared with other studies. Also the gross motor development curves and the maximum GMFM score derived from this study were compared with the Palisano's report and the Rosenbaum's report. Results: Based on a total of 494 GMFM assessments provided by this study, the 5 distinct motor development curves and the maximum GMFM score were created. These observations is corresponding with the Palisano's and the Rosenbaum`s Development curves. Conclusion: The 5 distinct motor development curves (GMFCS) that were created by Palisano's and Rosenbaum's study is useful in Korea, providing parents and clinicians with a means to plan interventions and to judge progress over time.

• Communications for Statistical Applications and Methods
• /
• 제19권5호
• /
• pp.655-662
• /
• 2012

Classification is an important research field in pattern recognition with high-dimensional predictors. The support vector machine(SVM) is a penalized feature selector and classifier. It is based on the hinge loss function, the non-convex penalty function, and the smoothly clipped absolute deviation(SCAD) suggested by Fan and Li (2001). We developed the algorithm for the multiclass SVM with the SCAD penalty function using the local quadratic approximation. For multiclass problems we compared the performance of the SVM with the $L_1$, $L_2$ penalty functions and the developed method.

• 한국지능시스템학회논문지
• /
• 제15권3호
• /
• pp.289-293
• /
• 2005

패턴 분류는 실세계의 객체를 표현한 다양한 형태의 패턴 정보를 추출하여, 이것이 어떤 부류(클래스)인가를 결정하는 것이다. 패턴 분류 기술은 데이터 마이닝, 산업 자동화나 업무자동화를 위한 컴퓨터 응용 소프트웨어 기술로서 현재 다양한 분야에서 활용되고 있다. 패턴 분류 기술의 최대 목표는 분류 성능 향상이며 이것을 위해 지난 40년간 많은 연구자들이 다양한 접근 방법들을 시도해 왔다. 주로 이용되는 단일 분류 방법들로는 패턴들의 확률적 추론에 기반한 베이즈 분류기, 결정 트리, 거리함수를 이용하는 방법, 신경망, 군집화 등이 있으나 대용량 다차원 데이터를 분석하기에는 효율적이지 못하다. 따라서 상호 보완적인 여러 분류기들을 사용해 결합을 통하여 성능 향상에 도움을 주고 있는 다중 분류기 시스템에 대한 연구가 활발하게 진행되고 있다. 본 논문에서는 다중 SVM(Support Vector Machine) 분류기에 관한 기존 연구의 문제점을 지적하고 새로운 모델을 제안한다. SVM을 다중 클래스 분류기로 확장하기 위해 일대다 정책을 기반으로 하여 각각의 SVM 출력값을 비선형 패턴을 갖는 신호로 간주하고 이를 신경망에 학습하여 최종 분류 성능 결과를 결합하는 모델인 BORSE(Bootstrap Resampling SVM by Ensemble)를 제안한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권12호
• /
• pp.5116-5134
• /
• 2015

In this paper, due to the network security defense is mainly static defense, a dynamic classification network security defense strategy model is proposed by analyzing the security situation of complex computer network. According to the network security impact parameters, eight security elements and classification standard are obtained. At the same time, the dynamic classification algorithm based on fuzzy theory is also presented. The experimental analysis results show that the proposed model and algorithm are feasible and effective. The model is a good way to solve a safety problem that the static defense cannot cope with tactics and lack of dynamic change.

• 대한의용생체공학회:의공학회지
• /
• 제16권2호
• /
• pp.157-166
• /
• 1995

The Electroencephalogram (EEG) and evoked potential (EP) t;ave widely been used for study of brain functions. The EEG and EP signals acquired from multi-channel electrodes placed on the head surface are often interfered by other relatively large physiological signals such as electromyogram (EMG) or electroculogram (EOG). Since these artifact-affected EEG signals degrade EEG mapping, the removal of the artifact-affected EEGs is one of the key elements in neuro-functional mapping. Conventionally this task has been carried out by human experts spending lots of examination time. In this paper a neural-network based classification is proposed to replace or to reduce human expert's efforts and time. From experiments, the neural-network based classification performs as good as human experts : variation of decisions between the neural network and human expert appears even smaller than that between human experts.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2012년도 추계학술대회
• /
• pp.881-886
• /
• 2012

The paper is facial expression recognition based on the inclined face geometrical information. In facial expression recognition, mouth has a key role in expressing emotions, in this paper the features is mainly based on the shapes of mouth, followed by eyes and eyebrows. This paper makes its efforts to disperse every feature values via the weighting function and proposes method of expression classification with excellent classification effects; the final recognition model has been constructed.

• 한국디지털정책학회:학술대회논문집
• /
• 한국디지털정책학회 2007년도 춘계학술대회
• /
• pp.79-87
• /
• 2007

This paper aims to offer a solution based on semantic document classification to improve e-Government utilization and efficiency for people using their own information retrieval system and linguistic expression Generally, semantic document classification method is an approach that classifies documents based on the diverse relationships between keywords in a document without fully describing hierarchial concepts between keywords. Our approach considers the deep meanings within the context of the document and radically enhances the information retrieval performance. Concept Weight Document Classification(CoWDC) method, which goes beyond using exist ing keyword and simple thesaurus/ontology methods by fully considering the concept hierarchy of various concepts is proposed, experimented, and evaluated. With the recognition that in order to verify the superiority of the semantic retrieval technology through test results of the CoWDC and efficiently integrate it into the e-Government, creation of a thesaurus, management of the operating system, expansion of the knowledge base and improvements in search service and accuracy at the national level were needed.