• The Journal of the Acoustical Society of Korea
• /
• v.30 no.8
• /
• pp.461-467
• /
• 2011

Accurate classification of input signals is the key prerequisite for variable bit-rate coding, which has been introduced in order to effectively utilize limited communication bandwidth. Especially, recent surge of multimedia services elevate the importance of speech/music classification. Among many speech/music classifier, the ones based on support vector machine (SVM) have a strong selling point, high classification accuracy, but their computational complexity and memory requirement hinder their way into actual implementations. Therefore, techniques that reduce the computational complexity and the memory requirement is inevitable, particularly for embedded systems. We first analyze implementation of an SVM-based classifier on embedded systems in terms of execution time and energy consumption, and then propose two techniques that alleviate the implementation requirements: One is a technique that removes support vectors that have insignificant contribution to the final classification, and the other is to skip processing some of input signals by virtue of strong correlations in speech/music frames. These are post-processing techniques that can work with any other optimization techniques applied during the training phase of SVM. With experiments, we validate the proposed algorithms from the perspectives of classification accuracy, execution time, and energy consumption.

• Journal of the Korean Society for Railway
• /
• v.19 no.5
• /
• pp.672-684
• /
• 2016

UIC Codes 714R & 715R recommend the use of line classifications and their usage in maintenance work by employing notional traffic loads. However, the classification has not been applied to local lines and, therefore, a new line classification system based on UIC 714R has been proposed in this study. For this, various classification models of UIC, Germany, and UK have been studied first and equivalent traffic loads based on Korail's report, as well as on train timetables, have been derived. The results of the classifications have been compared with those of major European countries and it has been shown that the proposed classification is equivalent to the average value in the European cases. The line classification can be fully utilized during the decision making process of maintenance work and will also be used to model the Reliability Centered Maintenance (RCM) in the future.

• Journal of the Korean Society for Library and Information Science
• /
• v.30 no.2
• /
• pp.149-165
• /
• 1996

Hong-sii Tokso-rok (홍씨독서록 or Hong's Annotated Bibliography of Korean and Chinise Book) is the only work on the history of Korean bibliographies that has the introductory notes to each class, that is description of the origin of subject fields, transition, and characteristics at the beginning of each class. This paper is aimed to examine the outline of the introductory description of class, to analyze the Four Category Classification System(사부분류법) devised by Hong Sok-chu, and to explain how the classes of Four Category Classification are set and ordered. This paper shows several characteristics in the idea of Hong's classification system. There characteristics were discovered by analyzing the content of each introduction of classes. The characteristics ale as follows First, classes are organize and arranged from the substantial problem to nonsubstantial ones. In other words, the greater the distance of the class from the substantial problem of Confucianism, the farther the order of the class will be found from the substantial problem. The order of classes is set by how the class is closed to the substantial problem in the same hierarchy. This principle is strictly applied to the Hong's classification system. Second, on the basis of democratic thought, he del·eloped the classification system. In other words, when he set up the priority of classes, he put emphasis on the democratism as a guideline. The organization of classes belong to the catagories of history (Sa-bu, 사부) and philosophy(Cha-bu, 자부) showed the application of this principle. Conclusively, this paper found that Hong did not randomly arrange the class older, but he set the class order with objective reasons and logic when he set the class order of arrangement.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.41 no.4
• /
• pp.160-170
• /
• 2018

This paper is an exploratory study on the success factors of Defence Quality Management System (DQMS) which is the certification system granted by the military for improving the quality of munitions. DQMS is established by adding military requirements to the ISO standard, thus, we especially focus on the additional requirements to figure out success key factors of DQMS certification. The 51 additional requirements of Korean Defense Specification (KDS) are empirically investigated from 67 companies that acquired DQMS certification. Firstly, we conduct an independent t-tests on 51 additional requirements of KDS 0050-900-3 to determine if there is a difference between an easily certified company and a hard-to-certify company, and obtain 8 requirements such as 'Internal propagation of performance', 'Preparation of documented work instructions', 'Work instructions in the workplace', 'Documentation of equipment management', 'Inventory management', 'Packaging and identification', 'Guarantee of access to internal audit result for customers', 'Notification to the customer for improper product.' Secondly, we carry out an factor analysis to the 51 additional requirements for classification, and figure out that 4 requirements among the 8 requirements above mentioned are grouped together in the same factor. The 4 requirements are 'Preparation of documented work instructions', 'Work instructions in the workplace', 'Packaging and identification', and 'Guarantee of access to internal audit result for customers.' The result of this paper will provide useful information to the company preparing for DQMS.

• Journal of Welding and Joining
• /
• v.16 no.3
• /
• pp.64-73
• /
• 1998

This paper is concerned with defects detection and evaluation of heat affected zone (HAZ) in austenitic stainless steel type 304 by ultrasonic wave and neural network. In experiment, the reflected ultrasonic defect signals from artificial defects (side hole, vertical hole, notch) of HAZ appears as beam distance of prove-defect, distance of probe-surface, depth of defect-surface on CRT. For defect classification simulation, neural network system was organized using total results of ultrasonic experiment. The organized neural network system was learned with the accuracy of 99%. Also it could be classified with the accuracy of 80% in side hole, and 100% in vertical hole, 90% in notch about ultrasonic pattern recognition. Simulation results of neural network agree fairly well with results of ultrasonic experiment. Thus were think that the constructed system (ultrasonic wave - neural network) in this work is useful for defects dection and classification such as holes and notches in HAZ of austenitic stainless steel 304.

• Journal of Information Science Theory and Practice
• /
• v.7 no.3
• /
• pp.21-30
• /
• 2019

Information about the formation and filling (in 2014 to 2016) of a terminological dictionary on electronics and radioengineering and collective work (in 2017 to 2018) with a data bank "TERMIN" is presented in this article. In purpose of creating an instrument of navigating the modern scientific-technical space a net of terms with set semantic links is described. This set is based on the analysis of terms' definitions (each term is checked for inclusion in the definitions of all other terms; the definitions were borrowed from reputable reference editions: encyclopedias, dictionaries, reference books). The created model of a system that consists of different information sources, in which it (information) is indexed by the terminology of Russian State Rubricator of Scientific and Technical Information rubrics and/or keywords, is described. There is an access for the search in all these sources in the system. Searching inquiries are referred to in the language of these rubrics or formulated by arbitrary terms. The system is to refer to information sources and give out relevant information. In accordance with this model, semantic links of various types, which allow expanding a search at different modalities of query, should be set among data bank terms. Obtained links will have to increase semantic matching, i.e., they can provide actual understanding of the meaning of the information that is being sought.

• Journal of Society of Preventive Korean Medicine
• /
• v.13 no.2
• /
• pp.115-128
• /
• 2009

Background : Currently, ontology research has led the trend of technical development in medical informatics area. For Korean medicine, the reference terminology should be developed to facilitate the ontology research. Objective : This research aimed to design the management system for traditional Korean medical terms. Method :We built the internet-based system in which multi-users work simultaneously by using the relational database system(SQL Server2005) and visual studio 2005. Result : By this system, researchers can collect, refine, and inspect Korean medicine terms efficiently, and the terms can be transcribed into synonym, Korean, Chinese, and simplified Chinese. It enables the terms be input into the system accurately and managed by its classification. Conclusion : We developed the concept groups and its hierarchy system for Korean medicine terms which provides the basis for ontology system.

• Journal of the Korea Safety Management & Science
• /
• v.11 no.2
• /
• pp.117-126
• /
• 2009

We develop methods for propagating and analyzing EPCI(Extended Process Capability Index) by using the error type that classifies into accuracy and precision. EPCI developed in this study can be applied to the three combined processes that consist of production, measurement and calibration. Little calibration work discusses while a great deal has been studied about SPC(Statistical Process Contol) and MSA(Measurement System Analysis). EPCI can be decomposed into three indexes such as PPCI(Production Process Capability Index), PPPI(Production Process Performance Index), MPCI(Measurement PCD, and CPCI(Calibration PCI). These indexs based on the type of error classification can be used with various statistical techniques and principles such as SPC control charts, ANOVA(Analysis of Variance), MSA Gage R&R, Additivity-of-Variance, and RSSM(Root Sum of Square Method). As the method proposed is simple, any engineer in charge of SPC. MSA and calibration can use efficientily in industries. Numerical examples are presentsed. We recommed that the indexes can be used in conjunction with evaluation criteria.

• International Journal of Computer Science & Network Security
• /
• v.23 no.8
• /
• pp.40-48
• /
• 2023

Speech Emotions recognition has become the active research theme in speech processing and in applications based on human-machine interaction. In this work, our system is a two-stage approach, namely feature extraction and classification engine. Firstly, two sets of feature are investigated which are: the first one is extracting only 13 Mel-frequency Cepstral Coefficient (MFCC) from emotional speech samples and the second one is applying features fusions between the three features: Zero Crossing Rate (ZCR), Teager Energy Operator (TEO), and Harmonic to Noise Rate (HNR) and MFCC features. Secondly, we use two types of classification techniques which are: the Support Vector Machines (SVM) and the k-Nearest Neighbor (k-NN) to show the performance between them. Besides that, we investigate the importance of the recent advances in machine learning including the deep kernel learning. A large set of experiments are conducted on Surrey Audio-Visual Expressed Emotion (SAVEE) dataset for seven emotions. The results of our experiments showed given good accuracy compared with the previous studies.

• International Journal of Computer Science & Network Security
• /
• v.24 no.1
• /
• pp.107-118
• /
• 2024

With the seamless growth of the technology, network usage requirements are expanding day by day. The majority of electronic devices are capable of communication, which strongly requires a secure and reliable network. Network-based intrusion detection systems (NIDS) is a new method for preventing and alerting computers and networks from attacks. Machine Learning is an emerging field that provides a variety of ways to implement effective network intrusion detection systems (NIDS). Bagging and Boosting are two ensemble ML techniques, renowned for better performance in the learning and classification process. In this paper, the study provides a detailed literature review of the past work done and proposed a novel ensemble approach to develop a NIDS system based on the voting method using bagging and boosting ensemble techniques. The test results demonstrate that the ensemble of bagging and boosting through voting exhibits the highest classification accuracy of 99.98% and a minimum false positive rate (FPR) on both datasets. Although the model building time is average which can be a tradeoff by processor speed.