• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.6 no.2
• /
• pp.149-158
• /
• 2005

In this paper, we have designed and implemented a portable e-tongue (electronic tongue) system using MACS (multi array chemical sensor) and PDA. The system embedded in PDA has merits such as comfortable user interface and data transfer by internet from on-site to remote computer. MACS was made up 7 electrodes (${NH_4}^+$, $Na^+$, $Cl^-$, ${NO_3}^-$, $K^+$, $Ca^{2+}$, $Na^+$, pH) and a reference electrode. For learning the system, we adapted the Levenberg-Marquardt algorithm based on the back-propagation, which could iteratively learned the pre-determined standard patterns, in e-tongue system. Conclusionally, the relationship between the standard patterns and unknown pattern can be easily analyzed. The e-tongue was applied to whiskeys and cognac (one high level whisky, one low level whiskey, two cognac) and 2 sample whiskeys for each standard patterns and unknown patterns. The relationship between the standard patterns and unknown patterns can be easily analyzed.

• Journal of KIISE:Databases
• /
• v.41 no.4
• /
• pp.242-248
• /
• 2014

With the rapidly growing amount of information represented in RDF format, efficient querying of RDF graph has become a fundamental challenge. SPARQL is one of the most widely used query languages for retrieving information from RDF dataset. SPARQL is not only simple in its syntax but also powerful in representation of graph pattern queries. However, users need to make a lot of efforts to understand the ontology schema of a dataset in order to compose a relevant SPARQL query. In this paper, we propose a graph query formulation and processing scheme based on ontology schema information which can be obtained by summarizing RDF graph. In the context of the proposed querying scheme, a user can interactively formulate the graph queries on the graphic user interface without making efforts to understand the ontology schema and even without learning SPARQL syntax. The graph query formulated by a user is transformed into a set of class paths, which are stored in a relational database and used as the constraint for search space reduction when the relational database executes the graph search operation. By executing the LUBM query 2, 8, and 9 over LUBM (10,0), it is shown that the proposed querying scheme returns the complete result set.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.37 no.6
• /
• pp.551-558
• /
• 2013

A support vector machine (SVM) is a very powerful classification algorithm when a set of training data, each marked as belonging to one of several categories, is given. Therefore, SVM techniques have been used as one of the diagnostic tools in machine learning as well as in pattern recognition. In this paper, we present the results of classifying ball bearing fault types and severities using SVM with an optimized feature set based on the minimum distance rule. A feature set as an input for SVM includes twelve time-domain and nine frequency-domain features that are extracted from the measured vibration signals and their decomposed details and approximations with discrete wavelet transform. The vibration signals were obtained from a test rig to simulate various bearing fault conditions.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.5
• /
• pp.99-106
• /
• 2015

This research is focused on the development of the proprietary database embedded in the OTT device, which is used for searching and indexing video contents, and also the development of the search algorithm in the form of the critical components of the interface application with the OTT's database to provide video query searching, such as remote control smartphone application. As the number of available channels has increased to anywhere from dozens to hundreds of channels, it has become increasingly difficult for the viewer to find programs they want to watch. To address this issue, content providers are now in need of methods to recommend programs catering to each viewer's preference. the present study aims provide of the algorithm which recommends contents of OTT program by analyzing personal watching pattern based on one's history.

• Journal of Sensor Science and Technology
• /
• v.8 no.6
• /
• pp.461-468
• /
• 1999

In this paper, we have implemented a gas recognition system for classification and identification of explosive gases such as methane, propane, and butane using a sensor array and an artificial neural network. Such explosive gases which can be usually detected in the oil factory and LPG pipeline are very dangerous for a human being. We analyzed the characteristics of a multi-dimensional sensor signals obtained from the nine sensors using the principal component analysis(PCA) technique. Also, we implemented a gas pattern recognizer using a multi-layer neural network with error back propagation learning algorithm, which can classify and identify the sorts of gases and concentrations for each gas. The simulation and experimental results show that the proposed gas recognition system is effective to identify the explosive gases. And also, we used DSP board(TMS320C31) to implement the proposed gas recognition system using the neural network for real time processing.

• Journal of KIISE:Software and Applications
• /
• v.32 no.12
• /
• pp.1260-1270
• /
• 2005

Support Vector Machines (SVM) is well known for a representative learner as one of the kernel methods. SVM which is based on the statistical learning theory shows good generalization performance and has been applied to various pattern recognition problems. However, SVM is basically to deal with a two-class classification problem, so we cannot solve directly a multi-class problem with a binary SVM. One-Per-Class (OPC) and All-Pairs have been applied to solve the face recognition problem, which is one of the multi-class problems, with SVM. The two methods above are ones of the output coding methods, a general approach for solving multi-class problem with multiple binary classifiers, which decomposes a complex multi-class problem into a set of binary problems and then reconstructs the outputs of binary classifiers for each binary problem. In this paper, we introduce the output coding methods as an approach for extending binary SVM to multi-class SVM and propose new output coding schemes based on the Error-Correcting Output Codes (ECOC) which is a dominant theoretical foundation of the output coding methods. From the experiment on the face recognition, we give empirical results on the properties of output coding methods including our proposed ones.

• Journal of Internet Computing and Services
• /
• v.19 no.3
• /
• pp.15-23
• /
• 2018

A code reuse attack is an attack technique that can execute arbitrary code without injecting code directly into the stack by combining executable code fragments existing in program memory and executing them continuously. ROP(Return-Oriented Programming) attack is typical type of code reuse attack and serveral defense techniques have been proposed to deal with this. However, since existing methods use Rule-based method to detect attacks based on specific rules, there is a limitation that ROP attacks that do not correspond to previously defined rules can not be detected. In this paper, we introduce a method to detect ROP attack by learning command pattern used in ROP attack code using RNN(Recurrent Neural Network). We also show that the proposed method effectively detects ROP attacks by measuring False Positive Ratio, False Negative Ratio, and Accuracy for normal code and ROP attack code discrimination.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2010.09b
• /
• pp.21-31
• /
• 2010

Although it is important to reflect the accurate information of the ground condition in the tunnel design, the analysis and design are conducted by limited information because it is very difficult to get it practically on considering various geography and geotechnical condition. So construction management of information concept is required to manage immediately on the field condition because it is very time-consuming to establish the countermeasure of underground reinforcement and the pattern change of Bo. Therefore, when construction is on tunnel area, examination of accurate safety and prediction of behavior is performed to overcomes the limit of predicting behavior by using Artificial Neural Network(ANN) in this study. Firstly, the field data was secured. Secondly, suitable structure was made on multi-layer perceptrons among the ANN. Thirdly, learning algorithm-propagated applies to ANN. The data for the learn of field application using ANN was used by considering impact factors, which influenced the behavior of tunnel, and performing credibility analysis. crown displacement, spring displacement, subsurfacement, and rock bolt axial force are predicted at the tunnel construction and on-site application was confirmed by using ANN from analyzing and comparing with measurement value of on-site. In this study, the data from Seoul Highway $\bigcirc\bigcirc$ tunnel section was applied to the ANN Theory, and the analysis on the investigate value and the reasoning for the value associated with field application was performed.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.3
• /
• pp.22-28
• /
• 2010

Adaboost is well known for a representative learner as one of the kernel methods. Adaboost which is based on the statistical learning theory shows good generalization performance and has been applied to various pattern recognition problems. However, Adaboost is basically to deal with a two-class classification problem, so we cannot solve directly a multi-class problem with Adaboost. One-Vs-All and Pair-Wise have been applied to solve the multi-class classification problem, which is one of the multi-class problems. The two methods above are ones of the output coding methods, a general approach for solving multi-class problem with multiple binary classifiers, which decomposes a complex multi-class problem into a set of binary problems and then reconstructs the outputs of binary classifiers for each binary problem. However, two methods cannot show good performance. In this paper, we propose the method to solve a multi-target classification problem by using radial basis function of Adaboost weak classifier.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.3
• /
• pp.29-40
• /
• 2020

Korean firms have been vigorously searching and exploring overseas market opportunities through export and overseas investment. As of end of 2019, there were more than 80,000 Korean overseas subsidiaries all over the world. With Korean overseas direct investment increasing recently, it became one of the important issues for overseas investors to be successful in the global market. There are a lot of studies on factors influencing the performance of overseas subsidiaries such as 'firm' and 'country' factors. This study empirically examines subsidiary performance determinants with 'industry architectures' by using a sample of 292 overseas Korean firm subsidiaries. Industry architectures are the stable but evolving sets of rules and roles through which labor is divided within a sector. This article considers how industry architectures shape success in international expansion. Industry architectures differ between countries, are not necessarily technologically determined, shape firms' capabilities and their competitive environment, and constitute a distinct level of analysis. We extract antecedents of related theory and empirically test its impact with a survey of Korean firms expanding in emerging economies. We would say this is the first study which tries to focus on industry architectures with the performance of Korean overseas subsidiaries. We find that separability and similarity of industry architectures across countries and localization of subsidiaries are robust and important predictors of success in international expansion. Our results suggest that industry architectures should be added to firm and country as an intermediate level of analysis that helps explain success in international expansion. While we established a pattern, much more remains to be done. We focus on the success of foreign operations, but we do not consider the broader benefits of going abroad, such as the learning or network effects that accrue at the level of the entire firm. The next obvious question is whether the results would differ in the developed market context. These we leave for future research to consider.