• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.4
• /
• pp.81-90
• /
• 1996

In this paper, we implemented a computerized pulse diagnosis system for the developement of pulse diagnosis algorithms and the generalization of the pulse diagnosis. The system consists of the hardware and software. The hardware detects pulse waves and inputs the waves into the computer system, while the software not only manages and analyzes the input pulse wave data but also privides the database. In order to clinically test the developed pulse diagnosis system, we applied the carotid-radial pulse diagnosis algorithm to the system. As the results, it is found that the genralizatio of the pulse diagnosis is possible.

• Journal of the Korea Safety Management & Science
• /
• v.5 no.2
• /
• pp.87-98
• /
• 2003

With the development of CNC(Computer Numerical Control) and communication technology, the connotation and functions of Distributed Numerical Control have been greatly enlarged. In this study, we develop and implement a Distributed Numerical Control system that has real time and multi-tasking operation capability for the machining cell with various NC(Numerical Control) and CNC machines. With the consideration of economy, generalization and extension, this system is interfaced with Shop Floor Control System, Machine Control System and Tool Preparation System using advanced networking method. In the implementation phase, we use the ORACLE DBMS (Database Management System) as the DBMS and Microsoft Visual C++ as the programming tools.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.77-88
• /
• 2021

Heart Failure represents a critical pathological case that is challenging to predict and discover at an early age, with a notable increase in morbidity and mortality. Machine Learning and Neural Network techniques play a crucial role in predicting heart attacks, diseases and more. These techniques give valuable perspectives for clinicians who may then adjust their diagnosis for each individual patient. This paper evaluated neural network models for heart attacks predictions. Several online learning methods were investigated to automatically and accurately predict heart attacks. The UCI dataset was used in this work to train and evaluate First Order and Second Order Online Learning methods; namely Backpropagation, Delta bar Delta, Levenberg Marquardt and QuickProp learning methods. An optimizer technique was also used to minimize the random noise in the database. A regularization concept was employed to further improve the generalization of the model. Results show that a three layers' NN model with a Backpropagation algorithm and Nadam optimizer achieved a promising accuracy for the heart attach prediction tasks.

• The KIPS Transactions:PartD
• /
• v.9D no.5
• /
• pp.819-826
• /
• 2002

The mobile vector map services for mobile devices, specially for PDA, are being popularized with the advent of KVM (Kilobytes Virtual Machine). However, the services for cellular phone have some problems for implementation, which are the restrictions of bandwidth, resource of the phone, etc. Thus, these restrictions force to develop new map database exclusively for cellular phone services. But, this approach needs high costs and could not support multi user view for a same area. Therefore, we suggest new client/server system for mobile vector map services, using existing general map database, and design and implement the system. The suggested system, which has minimized functions at client and processes map simplification works at sever, makes it possible to serve mobile vector map without developing new map database only for cellular phone.

• Journal of Korean Society for Geospatial Information Science
• /
• v.10 no.1 s.19
• /
• pp.109-123
• /
• 2002

The purpose of this study construct water-facility Database and to develop its management system using GSIS. We used route system and event table that are data model in GSIS for the construction of water valves on water facility pipe. A water-facility management system using GSIS was developed and was actually applied for the city of Yosu and compared with conventional water-facility management systems in terms of workability, information searching cost, and database management efficiency. The system has also been compared with those of two other local governments in terms of functions, process time, and conveniences. The generalization of the work model for water-facility system was also attempted based on the data collected from three local governments and it was revised by the work model of the wide area water-facility system. The generalized work model in this study was compared with other two local governments' work models. The generalized work model will enable local governments to develop more efficient and economical water-facility management system by GSIS in future.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.12
• /
• pp.4008-4023
• /
• 2022

Since machine learning was introduced into cross-site scripting (XSS) attack detection, many researchers have conducted related studies and achieved significant results, such as saving time and labor costs by not maintaining a rule database, which is required by traditional XSS attack detection methods. However, this topic came across some problems, such as poor generalization ability, significant false negative rate (FNR) and false positive rate (FPR). Moreover, the automatic clustering property of graph convolutional networks (GCN) has attracted the attention of researchers. In the field of natural language process (NLP), the results of graph embedding based on GCN are automatically clustered in space without any training, which means that text data can be classified just by the embedding process based on GCN. Previously, other methods required training with the help of labeled data after embedding to complete data classification. With the help of the GCN auto-clustering feature and labeled data, this research proposes an approach to detect XSS attacks (called GCNXSS) to mine the dependencies between the units that constitute an XSS payload. First, GCNXSS transforms a URL into a word homogeneous graph based on word co-occurrence relationships. Then, GCNXSS inputs the graph into the GCN model for graph embedding and gets the classification results. Experimental results show that GCNXSS achieved successful results with accuracy, precision, recall, F1-score, FNR, FPR, and predicted time scores of 99.97%, 99.75%, 99.97%, 99.86%, 0.03%, 0.03%, and 0.0461ms. Compared with existing methods, GCNXSS has a lower FNR and FPR with stronger generalization ability.

• Proceedings of the Korea Database Society Conference
• /
• 1997.10a
• /
• pp.483-503
• /
• 1997

One important question that arises when dealing with temporal databases in context of object-oriented systems is the method that associates time with attributes relationship semantics. Results of previous work about attribute versioning, particularity extending flat(First Normal Form: FNF) or nested(Non-First Normal Form: NFNF) relational models. are not applicable to temporal object-oriented databases. This is because object-oriented models provide more powerful constructs than traditional models for structuring complex objects. Therefore, this paper presents an formal approach for incorporating temporal extension to object-oriented databases. Our goal in this paper is to study temporal object-oriented database representation according to generalization, aggregation and association among objects. We define tile concepts of attribute versioning in temporal object-oriented model, and we concentrate on the representation of temporal relationship among objects. Another contribution of this paper is to specify time constraints on relationship semantics and analyze our model based on representation criteria. By means of formalizing tile temporal object oriented model, this paper can not only provide tile robust operating functions that design algebraic operators, but also entrance the reuse of modules.

• Journal of Korean Society on Water Environment
• /
• v.32 no.6
• /
• pp.511-519
• /
• 2016

National Institute of Environmental Research (NIER) has developed the Korean Reach File (KRF) for scientific and systematic analysis of variables related to water quality, pollutant sources and aquatic ecosystems in consideration of steam reach networks. The KRF provides a new framework for data production, storage, management and analysis for water related variables in relation to spatial characteristics, connections, and topologies of stream reaches. However, the current version of KRF (ver.2) has limited applicability because its nodes include not only the stream points based on topological characteristics but also those based on water quality monitoring stations, which may undermine its generality. In this study, a new version of KRF (ver.3) was designed and established to overcome the weak point of version 2. The version 3 is a generalization of the old KRF graphic data and it integrates the attribute data while separating it from the graphic data to minimize additional work that is needed for data association and search. We tested the KRF (ver.3) on actual cases and convenience and adaptability for each application was verified. Further research should focus on developing a database link model and real-world applications that are targeted to process event data.

• Wind and Structures
• /
• v.36 no.6
• /
• pp.393-404
• /
• 2023

Skewness and kurtosis are important higher-order statistics for simulating non-Gaussian wind pressure series on low-rise buildings, but their predictions are less studied in comparison with those of the low order statistics as mean and rms. The distribution gradients of skewness and kurtosis on roofs are evidently higher than those of mean and rms, which increases their prediction difficulty. The conventional artificial neural networks (ANNs) used for predicting mean and rms show unsatisfactory accuracy in predicting skewness and kurtosis owing to the limited capacity of shallow learning of ANNs. In this work, the deep neural networks (DNNs) model with the ability of deep learning is introduced to predict the skewness and kurtosis on a low-rise building. For obtaining the optimal generalization of the DNNs model, the hyper parameters are automatically determined by Bayesian Optimization (BO). Moreover, for providing a benchmark for future studies on predicting higher order statistics, the data sets for training and testing the DNNs model are extracted from the internationally open NIST-UWO database, and the prediction errors of all taps are comprehensively quantified by various error metrices. The results show that the prediction accuracy in this study is apparently better than that in the literature, since the correlation coefficient between the predicted and experimental results is 0.99 and 0.75 in this paper and the literature respectively. In the untrained cornering wind direction, the distributions of skewness and kurtosis are well captured by DNNs on the whole building including the roof corner with strong non-normality, and the correlation coefficients between the predicted and experimental results are 0.99 and 0.95 for skewness and kurtosis respectively.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.33 no.6
• /
• pp.303-310
• /
• 2019

In order to search for the causes of having difficulty with the scientification task of the Korean medicine, the definition and conception of science were reviewed first and then formalization of reasoning scheme and a practical method of scientification were proposed. Science in its definition is meant by foundation of method and system for production of scientific knowledge not by knowledge of science itself. The formation of science is composed of complex processes including not only scientific knowledge but also politicosocial output containing activity of scientist society, spreading of social value and intercommunication. The production of scientific knowledge of Korean medicine is begun from logicality of the differential diagnosis and treatment theory through abductive verification of analogical inference by yinyang and 5 phase theory. For the commensurability between the various heterogenic theories within Korean medicine, the scientific activity of collecting, compiling, analyzing, distributing, and discussing the significant knowledge gained through abductive verification in the experiment and clinical process should be formed broadly. Based on these knowledge database, organization of scientist society with Korean medicine, life science, medical engineering, social expansion and generalization of pattern conception, and then social propagation and contribution for national health should be driven forward.