• 산업경영시스템학회지
• /
• 제19권37호
• /
• pp.41-51
• /
• 1996

This paper evaluates the clustering performance of a neural network and a statistical method. Algorithms which are used in this paper are the GLVQ(Generalized Learning vector Quantization) for a neural method and the k-means algorithm fer a statistical clustering method. For comparison of two methods, we calculate the Rand's c statistics. As a result, the mean of c value obtained with the GLVQ is higher than that obtained with the k-means algorithm, while standard deviation of c value is lower. Experimental data sets were the Fisher's IRIS data and patterns extracted from handwritten numerals.

• International Journal of Computer Science & Network Security
• /
• 제21권2호
• /
• pp.9-13
• /
• 2021

Type 2 diabetes mellitus (T2D) is a complex diabetes disease that is caused by high blood sugar, insulin resistance, and a relative lack of insulin. Many studies are trying to predict variant genes that causes this disease by using a sample disease model. In this paper we predict diabetic and normal persons by using fisher score feature selection, chi-2 feature selection and Logistic Regression supervised learning algorithm with best accuracy of 90.23%.

• 한국컴퓨터정보학회논문지
• /
• 제10권5호
• /
• pp.77-86
• /
• 2005

본 논문에서는 분류오차를 추출하고 학습하여 분류성능을 개선하는 LVQ 학습 알고리즘을 설계하였다. 제안된 LVQ학습 알고리즘은 초기기준백터의 학습을 위해 SOM을 이용하고, LVQ 출력뉴런의 부류지정을 위하여 out-star 학습법을 사용하는 학습네트워크이다. 분류오차가 발생되는 패턴백터로 추출하기 위하여 오차유발조건을 제안하였고, 이 조건을 이용하여 분류오차를 유발시키는 입력패턴벡터로 구성되는 패턴백터공간을 구성하여 분류오차가 발생되는 패턴백터를 학습시키므로 분류오차수를 감소시키고, 패턴분류성능을 개선하였다. 제안된 학습알고리즘의 성능을 검증하기 위하여 Fisher의 Iris 데이터와 EMG 데이터를 학습백터 및 시험 백터로 사용하여 시뮬레이션 하였고, 제안된 학습방식의 분류 성능은 기존의 LVQ와 비교되어 기존의 학습방식보다 우수한 분류성공률을 확인하였다.

• 전자공학회논문지
• /
• 제52권8호
• /
• pp.117-125
• /
• 2015

본 논문은 '좌', '우' 방향 제어를 위해 취득된 EEG(Electroencephalogram) 신호 기반 분류 알고리즘과 EEG 센서, Labview, DAQ, Matlab, 주행로봇으로 구성된 방향 제어 시스템을 제안한다. 제안된 알고리즘은 DWT(Discrete Wavelet Transform)로 추출된 주파수대역 정보를 특징으로 이용하며, Fishers score를 이용하여 변별력이 높은 주파수 대역의 특징을 선별한다. 또한, SVM (Support Vector Machine)을 이용하여 분류 성능이 최고가 되는 특징벡터의 조합을 제안하고, 잘못된 판정에 의한 오동작을 방지하기 위한 MLD(Maximum Likelihood Decision) 기반의 판정보류 알고리즘도 제안한다. 제안된 알고리즘에 의해 선택된 4개의 특징벡터는 국제 표준 전극 배치법에 따른 P8 채널의 d2(16-32Hz), d5(2-4Hz) 주파수 대역의 전압의 절대 값 평균과 표준편차이다. SVM 분류기로 실험한 결과 98.75%의 정확도와 1.25%의 오류율 성능을 보였다. 또한, 오류 확률 70%를 판정 보류로 규정할 경우, 제안된 알고리즘은 인식률 95.63%의 정확도와 오류율 0%을 보였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권9호
• /
• pp.3762-3781
• /
• 2020

Infrared pedestrian target recognition is a vital problem of significant interest in computer vision. In this work, a novel infrared pedestrian target recognition method that uses heterogeneous features with multi-kernel transfer learning is proposed. Firstly, to exploit the characteristics of infrared pedestrian targets fully, a novel multi-scale monogenic filtering-based completed local binary pattern descriptor, referred to as MSMF-CLBP, is designed to extract the texture information, and then an improved histogram of oriented gradient-fisher vector descriptor, referred to as HOG-FV, is proposed to extract the shape information. Second, to enrich the semantic content of feature expression, these two heterogeneous features are integrated to get more complete representation for infrared pedestrian targets. Third, to overcome the defects, such as poor generalization, scarcity of tagged infrared samples, distributional and semantic deviations between the training and testing samples, of the state-of-the-art classifiers, an effective multi-kernel transfer learning classifier called MK-TrAdaBoost is designed. Experimental results show that the proposed method outperforms many state-of-the-art recognition approaches for infrared pedestrian targets.

• Smart Structures and Systems
• /
• 제25권2호
• /
• pp.169-181
• /
• 2020

Sensor placement is a crucial aspect of bridge health monitoring (BHM) dedicated to accurately estimate and locate structural damages. In addressing this goal, a sensor placement framework based on the deflection influence line (DIL) analysis is here proposed, for the optimal design of damage detection-oriented BHM system. In order to improve damage detection accuracy, we explore the change of global stiffness matrix, damage coefficient matrix and DIL vector caused by structural damage, and thus develop a novel sensor placement framework based on the Fisher information matrix. Our approach seeks to determine the contribution of each sensing node to damage detection, and adopts a distance correction coefficient to eliminate the information redundancy among sensors. The proposed damage detection-oriented optimal sensor placement (OSP) method is verified by two examples: (1) a numerically simulated three-span continuous beam, and (2) the Pinghu bridge which has existing real damage conditions. These two examples verify the performance of the distance corrected damage sensitivity of influence line (DSIL) method in significantly higher contribution to damage detection and lower information redundancy, and demonstrate the proposed OSP framework can be potentially employed in BHM practices.

• Smart Structures and Systems
• /
• 제4권5호
• /
• pp.629-640
• /
• 2008

A new algorithm is proposed to determine optimal accelerometer locations (OAL) when a structure is identified by frequency domain system identification (SI) method. As a result, a guideline is presented for selecting OAL which can reflect modal response of a structure properly. The guideline is to provide a minimum number of necessary accelerometers with the variation in the number of measurable target modes. To determine OAL for SI applications effectively, the modal sensitivity effective independence distribution vector (MS-EIDV) is developed with the likelihood function of measurements. By maximizing the likelihood of the occurrence of the measurements relative to the predictions, Fisher Information Matrix (FIM) is derived as a function of mode shape sensitivity. This paper also proposes a statistical approach in determining the structural parameters with a presumed parameter error which reflects the epistemic paradox between the determination of OAL and the application of a SI scheme. Numerical simulations have been carried out to examine the proposed OAL algorithm. A two-span multi-girder bridge and a two-span truss bridge were used for the simulation studies. To overcome a rank deficiency frequently occurred in inverting a FIM, the singular value decomposition scheme has been applied.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1243-1263
• /
• 2018

The two-stage linear discrimination analysis (TSLDA) is a feature extraction technique to solve the small size sample problem in the field of image recognition. The TSLDA has retained all subspace information of the between-class scatter and within-class scatter. However, the feature information in the four subspaces may not be entirely beneficial for classification, and the regularization procedure for eliminating singular metrics in TSLDA has higher time complexity. In order to address these drawbacks, this paper proposes an improved two-stage linear discriminant analysis (Improved TSLDA). The Improved TSLDA proposes a selection and compression method to extract superior feature information from the four subspaces to constitute optimal projection space, where it defines a single Fisher criterion to measure the importance of single feature vector. Meanwhile, Improved TSLDA also applies an approximation matrix method to eliminate the singular matrices and reduce its time complexity. This paper presents comparative experiments on five face databases and one handwritten digit database to validate the effectiveness of the Improved TSLDA.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4502-4518
• /
• 2015

Fisher linear discriminant analysis (LDA) is one of the most popular projection techniques for feature extraction and has been widely applied in face recognition. However, it cannot be used when encountering the single sample per person problem (SSPP) because the intra-class variations cannot be evaluated. In this paper, we propose a novel method called local similarity based linear discriminant analysis (LS_LDA) to solve this problem. Motivated by the "divide-conquer" strategy, we first divide the face into local blocks, and classify each local block, and then integrate all the classification results to make final decision. To make LDA feasible for SSPP problem, we further divide each block into overlapped patches and assume that these patches are from the same class. To improve the robustness of LS_LDA to outliers, we further propose local similarity based median discriminant analysis (LS_MDA), which uses class median vector to estimate the class population mean in LDA modeling. Experimental results on three popular databases show that our methods not only generalize well SSPP problem but also have strong robustness to expression, illumination, occlusion and time variation.

• ETRI Journal
• /
• 제44권3호
• /
• pp.462-475
• /
• 2022

Although researchers have proposed numerous techniques for speech emotion recognition, its performance remains unsatisfactory in many application scenarios. In this study, we propose a speech emotion recognition model based on a genetic algorithm (GA)-decision tree (DT) fusion of deep and acoustic features. To more comprehensively express speech emotional information, first, frame-level deep and acoustic features are extracted from a speech signal. Next, five kinds of statistic variables of these features are calculated to obtain utterance-level features. The Fisher feature selection criterion is employed to select high-performance features, removing redundant information. In the feature fusion stage, the GA is is used to adaptively search for the best feature fusion weight. Finally, using the fused feature, the proposed speech emotion recognition model based on a DT support vector machine model is realized. Experimental results on the Berlin speech emotion database and the Chinese emotion speech database indicate that the proposed model outperforms an average weight fusion method.