• The Journal of Korean Medicine Ophthalmology and Otolaryngology and Dermatology
• /
• v.17 no.3
• /
• pp.1-7
• /
• 2004

Objectives: In order to establish a base for proper treatment and management of patients with dermal problems through correct diagnosis, I considered the naming rule for dermatology in Oriental Medicine, referring to the dermatology literature compared to western medicine. In addition, this paper examines the characteristic classification of dermatology. Methods: I examined the naming rule of dermatology in Oriental Medicine and then compared the disease names in Oriental and Western medicine and the characteristic classification of dermatology referred to the records. Results: The dermal diseases have been named according to their colors and morphologies, causes, progress of symptoms, recurrent sites, the character of distribution, recurrent seasons, ages, the character of patients' jobs and locations. Sometimes some have been named by referring to their main morphologies, sites, causes, colors and seasons synthetically. However it was found some names for dermal diseases, even though the same diseases, had been named differently according to for example: historical times, condition of locations and the quality of doctors whose process of naming developed and changed over time. The relationship between Oriental and Western medicine of each name for dermal diseases is basically divided into 5 types: same names - same diseases; same names but different diseases; same diseases but different names; one disease with multiple names; and one name with multiple diseases. Considering the methods of classification, these were generally achieved according to their places of origin. It is a method unique to Oriental medicine that we classified some dermal diseases into 疥, 癬, 瘡, 風, 丹, 疱, 疹, 癰, 痘, 疽 and so on and it is very easy to diagnose which part they belong to. This was classified by putting first the causes of diseases; for instance: viruses, bacteria, fungi. Sometimes, however there was a problem, connected to the classification of morphology. Conclusions: I suggest that we need to unify and refine dermatological terms in Oriental Medicine in order to establish a base for proper treatment and management of patients with dermal problems through correct diagnoses.

• Geomechanics and Engineering
• /
• v.37 no.5
• /
• pp.475-498
• /
• 2024

The prediction of the susceptibility of soil to liquefaction using a limited set of parameters, particularly when dealing with highly unbalanced databases is a challenging problem. The current study focuses on different ensemble learning classification algorithms using highly unbalanced databases of results from in-situ tests; standard penetration test (SPT), shear wave velocity (V_s) test, and cone penetration test (CPT). The input parameters for these datasets consist of earthquake intensity parameters, strong ground motion parameters, and in-situ soil testing parameters. liquefaction index serving as the binary output parameter. After a rigorous comparison with existing literature, extreme gradient boosting (XGBoost), bagging, and random forest (RF) emerge as the most efficient models for liquefaction instance classification across different datasets. Notably, for SPT and V_s-based models, XGBoost exhibits superior performance, followed by Light gradient boosting machine (LightGBM) and Bagging, while for CPT-based models, Bagging ranks highest, followed by Gradient boosting and random forest, with CPT-based models demonstrating lower G_mean(error), rendering them preferable for soil liquefaction susceptibility prediction. Key parameters influencing model performance include internal friction angle of soil (ϕ) and percentage of fines less than 75 µ (F₇₅) for SPT and V_s data and normalized average cone tip resistance (q_c) and peak horizontal ground acceleration (a_max) for CPT data. It was also observed that the addition of V_s measurement to SPT data increased the efficiency of the prediction in comparison to only SPT data. Furthermore, to enhance usability, a graphical user interface (GUI) for seamless classification operations based on provided input parameters was proposed.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.6 no.2
• /
• pp.13-21
• /
• 1996

The necessity and usefulness of higher-order neural networks have been well-known since early days of neurocomputing. However the explosive number of terms has hampered the design and training of such networks. In this paper we present an evolutionary learning method for efficiently constructing problem-specific higher-order neural models. The crux of the method is the neural tree representation employing both sigma and pi units, in combination with the use of an MDL-based fitness function for learning minimal models. We provide experimental results in classification and prediction problems which demonstrate the effectiveness of the method. I. Introduction topology employs one hidden layer with full connectivity between neighboring layers. This structure has One of the most popular neural network models been very successful for many applications. However, used for supervised learning applications has been the they have some weaknesses. For instance, the fully mutilayer feedforward network. A commonly adopted connected structure is not necessarily a good topology unless the task contains a good predictor for the full *d*dWs %BH%W* input space.

• Journal of Korea Multimedia Society
• /
• v.22 no.11
• /
• pp.1251-1258
• /
• 2019

Intelligent tutoring system enables users to effectively learn by utilizing various artificial intelligence techniques. For instance, it can recommend a proper curriculum or learning method to individual users based on their learning history. To do this effectively, user's characteristics need to be analyzed and classified based on various aspects such as interest, learning ability, and personality. Even though data labeled by the characteristics are required for more accurate classification, it is not easy to acquire enough amount of labeled data due to the labeling cost. On the other hand, unlabeled data should not need labeling process to make a large number of unlabeled data be collected and utilized. In this paper, we propose a semi-supervised learning method based on feedback variational auto-encoder(FVAE), which uses both labeled data and unlabeled data. FVAE is a variation of variational auto-encoder(VAE), where a multi-layer perceptron is added for giving feedback. Using unlabeled data, we train FVAE and fetch the encoder of FVAE. And then, we extract features from labeled data by using the encoder and train classifiers with the extracted features. In the experiments, we proved that FVAE-based semi-supervised learning was superior to VAE-based method in terms with accuracy and F1 score.

• Korean Journal of Oriental Medicine
• /
• v.15 no.2
• /
• pp.93-100
• /
• 2009

Recently, there have been emerging research activities in classifying the Sasang constitution objectively by machine-based methods. The concordance rate of the classification by the pulse wave features was, however, only around 50% up to now. In this paper, we introduce a novel classification algorithm that can promote the accuracy substantially at the expense of the non-classifiable subgroup as a byproduct. For instance, with the pulse wave features alone, we show that female/male subject group in their 20s can be classified into the Sasang constitution group with the concordance rate of 68.4%/65.5% for a subgroup of 57/29(31%/15%) subjects out of 184/195, by leaving the other subjects as the non-classifiable group. Next, we show that the pulse diagnosis has been used only as a supportive tool in determining one's constitution, and consequently the accuracy of the concordance ratio by the pulse wave features alone cannot exceed a finite value, which we estimate to be about 60%.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.358-370
• /
• 2019

The main aim of this study is to select the optimal set of genes from microarray cancer datasets that contribute to the prediction of specific cancer types. This study proposes the enhancement of the feature selection filter algorithm based on Joe's normalized mutual information and its use for gene selection. The proposed algorithm is implemented and evaluated on seven benchmark microarray cancer datasets, namely, central nervous system, leukemia (binary), leukemia (3 class), leukemia (4 class), lymphoma, mixed lineage leukemia, and small round blue cell tumor, using five well-known classifiers, including the naive Bayes, radial basis function network, instance-based classifier, decision-based table, and decision tree. An average increase in the prediction accuracy of 5.1% is observed on all seven datasets averaged over all five classifiers. The average reduction in training time is 2.86 seconds. The performance of the proposed method is also compared with those of three other popular mutual information-based feature selection filters, namely, information gain, gain ratio, and symmetric uncertainty. The results are impressive when all five classifiers are used on all the datasets.

• 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.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.504-509
• /
• 2008

Support Vector Machine (SVM) is one of powerful learning machine and has been applied to varying task with generally acceptable performance. The success of SVM for classification tasks in one domain is affected by features which represent the instance of specific class. Given the representative and discriminative features, SVM learning will give good generalization and consequently we can obtain good classifier. In this paper, we will assess the problem of feature choices for human detection tasks and measure the performance of each feature. Here we will consider HOG-family feature. As a natural extension of SVM, we combine SVM with Principal Component Analysis (PCA) to reduce dimension of features while retaining most of discriminative feature vectors.

• Journal of Korean Library and Information Science Society
• /
• v.20
• /
• pp.155-186
• /
• 1993

This study aims at defining the concept of OPAC, tracing of research trends and development of it, and clarifying various methods of subject access and patterns of bibliographic searching in OPAC as well as strategies for improving to implement online catalogs. Although OPAC is so far the most user-friendly bibliographic searching method, there are still a lot of issues including online subject access in particular. Therefore a more effective and useful method for computer end-users have to be developed, for instance, improvement of an indexing system, a n.0, pplication of the classification system as a searching tool, a new design and construction of standardized thesaurus, betterment of user interface, introducing of expert system into bibliographic searching, establishment of subject authority file. Those would raise the success-rate of users seeking bibliographic information in the library catalogs. Korean libraries are in very early stage of OPAC implementation so that every efforts and concerns to improve strategies and techniques for subject access to OPAC have been strongly asked.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.41 no.4
• /
• pp.433-441
• /
• 1992

This paper describes an algorithm that can efficiently analyze the ECG signal using frame knowledge representation technique. Input to the analysis process is a set of significant points which have been extracted from an original sampled signal(lead II) by the syntactic peak recognition algorithm. The hierarchical property of ECG signal is represented by hierarchical AND/OR graph. The semantic information and constraints of the ECG signal are desctibed by frame. As the control mechanism for labeling points, the search mechanism with the mixed paradigms of data-driven and model driven hypothesis formation, scoring function, hypothesis modification network and instance inheritance are used. We used the CSE database in order to evaluate the performance of the proposed algorithm.