• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.1
• /
• pp.112-127
• /
• 2002

A useful pattern is a pattern that contributes much to learning. For a classification problem those patterns near the class boundary surfaces carry more information to the classifier. For a regression problem the ones near the estimated surface carry more information. In both cases, the usefulness is defined only for those patterns either without error or with negligible error. Using only the useful patterns gives several benefits. First, computational complexity in memory and time for learning is decreased. Second, overfitting is avoided even when the learner is over-sized. Third, learning results in more stable learners. In this paper, we propose a pattern 'utility index' that measures the utility of an individual pattern. The utility index is based on the bias and variance of a pattern trained by a network ensemble. In classification, the pattern with a low bias and a high variance gets a high score. In regression, on the other hand, the one with a low bias and a low variance gets a high score. Based on the distribution of the utility index, the original training set is divided into a high-score group and a low-score group. Only the high-score group is then used for training. The proposed method is tested on synthetic and real-world benchmark datasets. The proposed approach gives a better or at least similar performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2667-2682
• /
• 2015

Due to the wide application of face recognition (FR) in information security, surveillance, access control and others, it has received significantly increased attention from both the academic and industrial communities during the past several decades. However, partial face occlusion is one of the most challenging problems in face recognition issue. In this paper, a novel method based on linear regression-based classification (LRC) algorithm is proposed to address this problem. After all images are downsampled and divided into several blocks, we exploit the evaluator of each block to determine the clear blocks of the test face image by using linear regression technique. Then, the remained uncontaminated blocks are utilized to partial occluded face recognition issue. Furthermore, an improved Distance-based Evidence Fusion approach is proposed to decide in favor of the class with average value of corresponding minimum distance. Since this occlusion removing process uses a simple linear regression approach, the completely computational cost approximately equals to LRC and much lower than sparse representation-based classification (SRC) and extended-SRC (eSRC). Based on the experimental results on both AR face database and extended Yale B face database, it demonstrates the effectiveness of the proposed method on issue of partial occluded face recognition and the performance is satisfactory. Through the comparison with the conventional methods (eigenface+NN, fisherfaces+NN) and the state-of-the-art methods (LRC, SRC and eSRC), the proposed method shows better performance and robustness.

• Journal of Korean Academy of Nursing
• /
• v.36 no.2
• /
• pp.217-226
• /
• 2006

Purpose: The purpose of this study was to identify nursing interventions performed by public health nurses in health centers. Method: Data was collected by the taxonomy of Nursing Intervention Classification(NIC 3rd: 486 nursing interventions) from 131 public health nurses in health centers and analyzed using descriptive statistics. Result: As its result, more than 50% of public health nurses performed 137 nursing interventions at least monthly. The most frequently used intervention class was 'activity and exercise management', followed by 'physical comfort promotion', 'community health promotion', 'life span care', 'coping assistance', 'Self care facilitation', 'information management', 'nutrition support', 'community risk management' and 'patient education'. One hundred twenty nursing interventions were rarely performed by 90% or more of the nurses. Most of them were the physical complex domain. Conclusion: In conclusion, 137 interventions were performed by public health nurses at least monthly. NIC is helpful to build a standardized language for public health nursing.

• Journal of Korean Society for Geospatial Information Science
• /
• v.13 no.4 s.34
• /
• pp.33-38
• /
• 2005

The goal of this study is to segment building points effectively utilizing scan-line characteristics of airborne laser scanner. Points are classified as to their altitude similarity and adjacency with other classified points, and point searching range for the classification is restricted within some number of scan-lines, preventing classification speed from lowering as the process goes on. Besides, we detected wrong discrimination of one object into more than two classes, then integrated them into a single class. Consequently we could discriminate points of each building from others, its annexes and none building points simultaneously.

• Proceedings of the KIEE Conference
• /
• 1997.11a
• /
• pp.600-602
• /
• 1997

In this study, a hybrid neural net consisting of an Adaptive LVQ(ALVQ) algorithm and MLP is proposed to perform speaker identification task. ALVQ is a new learning procedure using adaptively feature vector sequence instead of only one feature vector in training codebooks initialized by LBG algorithm and the optimization criterion of this method is consistent with the speaker classification decision rule. ALVQ aims at providing a compressed, geometrically consistent data representation. It is fit to cover irregular data distributions and computes the distance of the input vector sequence from its nodes. On the other hand, MLP aim at a data representation to fit to discriminate patterns belonging to different classes. It has been shown that MLP nets can approximate Bayesian "optimal" classifiers with high precision, and their output values can be related a-posteriori class probabilities. The different characteristics of these neural models make it possible to devise hybrid neural net systems, consisting of classification modules based on these two different philosophies. The proposed method is compared with LBG algorithm, LVQ algorithm and MLP for performance.

• Proceedings of the Korean Society of Applied Pharmacology
• /
• 1993.04a
• /
• pp.53-53
• /
• 1993

In considering the mechanism of action of the calcium antagonists, it is important to realize that there are three distinct receptor types and that the new classification divides these three drugs as members of the dihydropyridine, phenylalkylamines and benzothiazipines, respectively. The World Health Organization as well as the International Union of Pharmacology and Cardiology have adopted this classification. Unlike every other class of drugs, such as the alpha and beta adrenergic blocking agents, diuretics, etc., the calcium antagonists need to be thought of as three distinct drug classes. The reason they share some, but not all of the pharmacological profile is that they all act at specific receptor domains present in one large protein of 165 daltons present in all excitable tissue. This protein along with several other subunits make up what is known as the voltage-dependent calcium channel (the so-called "L"type, L-VDCC). The mechanism of action of the three drugs involve first a specfic binding and then an inhibition of the movement of calcium into the cell Some of these drugs, such as diltiazem, may have other interesting intracellular effects perhaps associated with protection of the mitochondria during ischemic insults. The nature of the receptor is being explored by molecular genetic techniques, and we have recently cloned two of the major subunits; some of the data will be presented.

• International conference on construction engineering and project management
• /
• 2011.02a
• /
• pp.176-181
• /
• 2011

A role of cost modeler is that of facilitating design process by the systematic application of cost factors so as to maintain sensible and economic relationships between cost, quantity, utility and appearance. These relationships help to achieve the client's requirements within an agreed budget. The purpose of this study is to develop a parametric cost estimating model for the early design stage by using the multi-dimensional system of OLAP (On-line Analytical Processing) based on the case of quantity data related to architectural design features. The parametric cost estimating models have been adopted to support decision making in the early design stage. These models typically use a similar instance or a pattern of historical case. In order to effectively use this type of data model, it is required to set data classification and prediction methods. One of the methods is to find the similar class in line with attribute selection measure in the multi-dimensional data model. Therefore, this research is to analyze the relevance attribute influenced by architectural design features with the subject of case-based quantity data used for the parametric cost estimating model. The relevance attributes can be analyzed by Analytical Characterization. It helps determine what attributes to be included in the OLAP multi-dimension.

• Journal of Chest Surgery
• /
• v.26 no.3
• /
• pp.191-195
• /
• 1993

Among 25 consecutive cases having undergone mitral valve surgery between March 1991 and June 1992 in Gill General Hospital, 11 patients[44%] who had undergone mitral valve reconstruction using prosthetic rings is evaluated and presented. Patients` mean age is 43 + 19 years[range:16-72], and they are consisted with 4 males and 7 females. Mitral valve insufficiency is due to degenerative disease in 6 cases[55%] and rheumatic disease in 5 patients[45%]. Carpentier`s functional classification I is 2 cases, II is 6 cases, and III is 2 cases. Surgical techniques include prosthetic ring annuloplasty[11 patients, 100%], chorda shortening[6, 55%], leaflet mobilization[4,36%], new chorda formation[2, 18%], chorda transposition[1, 9%] commissurotomy[3, 27%], and papillary muscle splitting[3, 27%]. Average number of mitral anatomic lesions per patient are 2.7 and we used average 2.8 procedures upon mitral valve apparatus per patient. There were no surgical mortality and no late valve related admission during the mean follow up period of 17 months. The mean functional class[NYHA] is 2.81 preoperatively and improved to 1.10 postoperatively. Doppler echocardiography showed much improvement from grade II MR [1 case], grade III MR [1 case] and 9 cases of grade IV MR to 6 cases of patients showed no MR, only trace MR in 4 cases, and grade I MR was found only in one patient with NYHA functional class II postoperatively. The postoperative mean mitral valve area is $2.10+0.28cm^2$. We conclude that mitral reconstruction is a predictable and stable operation.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1134-1139
• /
• 2019

One of the most important applications of computer vision algorithms is the detection of threat objects in x-ray security images. However, in the practical setting, this task is complicated by two properties inherent to the dataset, namely, the problem of class imbalance and visual complexity. In our previous work, we resolved the class imbalance problem by using a GAN-based anomaly detection to balance out the bias induced by training a classification model on a non-practical dataset. In this paper, we propose a new method to alleviate the visual complexity problem by using a KNN-based automatic cropping algorithm to remove distracting and irrelevant information from the x-ray images. We use the cropped images as inputs to our current model. Empirical results show substantial improvement to our model, e.g. about 3% in the practical dataset, thus further outperforming previous approaches, which is very critical for security-based applications.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.3
• /
• 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.