• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.753-754
• /
• 2018

최근 인공지능에 대한 관심이 증가하고 관련 연구가 활발히 진행됨에 따라, 기존 연구분야에도 이를 적용하고자 하는 시도가 증가하고 있다. 본 연구진도 한글 글씨를 인식하기 위해 기계학습을 적용하고자 하며, 그에 따라 본 연구에서는 초기 연구로서 사용자 필기 데이터를 수집하기 위한 안드로이드용 앱을 개발하였다. 최종 대상이 한글 공부를 시작하는 유아로 선정하였으므로, 그에 적절하게 학습 앱의 Activity를 구성하였다. 입력한 한글 데이터 분만 아니라 하나의 글자에 대한 초성, 중성, 종성별로 데이터를 별도로 수집하여 추후 활용할 수 있게 구성하였다. 즉, 학습과정에서 발생한 데이터는 이미지와 이벤트 두 가지 모두 저장하여 추후 최종 연구에 활용하고자 하였다.

• Journal of information and communication convergence engineering
• /
• v.22 no.1
• /
• pp.80-87
• /
• 2024

The Manchu language holds historical significance, but a complete dataset of Manchu script letters for training optical character recognition machine-learning models is currently unavailable. Therefore, this paper describes the process of creating a robust dataset of extracted Manchu script letters. Rather than performing automatic letter segmentation based on whitespace or the thickness of the central word stem, an image of the Manchu script was manually inspected, and one copy of the desired letter was selected as a region of interest. This selected region of interest was used as a template to match all other occurrences of the same letter within the Manchu script image. Although the dataset in this study contained only 4,000 images of five Manchu script letters, these letters were collected from twenty-eight writing styles. A full dataset of Manchu letters is expected to be obtained through this process. The collected dataset was normalized and trained using a simple convolutional neural network to verify its effectiveness.

• The Journal of Korean Association of Computer Education
• /
• v.23 no.2
• /
• pp.53-64
• /
• 2020

As online learning forms centered on video lectures become more common and constantly increasing, the video-based learning environment applying various educational methods is also changing and developing to enhance learning effectiveness. Learner's log data has emerged for measuring the effectiveness of education in the online learning environment, and various analysis methods of log data are important for learner's customized learning prescriptions. To this end, the study analyzed learner behavior data and predictions of achievement by machine learning in video-based learning environments. As a result, interactive behaviors such as video navigation and comment writing, and learner-led learning behaviors predicted achievement in common in each model. Based on the results, the study provided implications for the design of the video learning environment.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.6
• /
• pp.637-646
• /
• 2022

In this paper the comparative study on the pronunciation of Vietnamese learners and Koreans for the Korean syllable final double consonants was performed. For many errors and the suggested teaching methods related to the pronunciation of the Korean syllable final double consonants that were investigated and analyzed through linguistic research the results of this study by using the analysis tools of speech signal processing were confirmed. Thus, we suggest the new educational method in this paper. Using SVM, which is widely used in machine learning of artificial intelligence the pronunciation of Vietnamese learners and that of Koreans were compared. Being able to obtain the decision hyperplane of the SVM means that Vietnamese learners' pronunciation of the Korean syllable final double consonants is quite different from that of Koreans. Otherwise their pronunciation are pretty similar each other. The new teaching method presented in this paper is not only composed of writing and listening but is included things such as the speech signal waveform in the time domain and its corresponding energy that can be visualized to the learners.

• Journal of KIISE
• /
• v.44 no.8
• /
• pp.767-773
• /
• 2017

In a distributed system environment, such as a cloud database, the tail latency needs to be kept short to ensure uniform quality of service. In this paper, through experiments on a Cassandra database, we show that long tail latency is caused by a lack of memory space because the database cannot receive any request until free space is reclaimed by writing the buffered data to the storage device. We observed that, since the performance of the storage device determines the amount of time required for writing the buffered data, the performance degradation of Solid State Drive (SSD) due to garbage collection results in a longer tail latency. We propose a garbage collection synchronization technique, called SyncGC, that simultaneously performs garbage collection in the java virtual machine and in the garbage collection in SSD concurrently, thus hiding garbage collection overheads in the SSD. Our evaluations on real SSDs show that SyncGC reduces the tail latency of $99.9^{th}$ and, $99.9^{th}-percentile$ by 31% and 36%, respectively.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.1
• /
• pp.263-271
• /
• 2016

Software requirement analysis is necessary for successful software development project. Specially, incomplete requirement is the most influential causes of software project failure. Incomplete requirement can bring late delay and over budget because of the misunderstanding and ambiguous criteria for project validation. Software requirement patterns can help writing more complete requirement. These can be a reference model and standards when author writing or validating software requirement. Furthermore, when a novice writes the software scenario, the requirement patterns can be one of the guideline. In this paper proposes an automatic approach to identifying software scenario patterns from various software scenarios. In this paper, we gathered 83 scenarios from eight industrial systems, and show how to extract 54 scenario patterns and how to find omitted action of the scenario using extracted patterns for the feasibility of the approach.

• Journal of Fashion Business
• /
• v.17 no.1
• /
• pp.131-144
• /
• 2013

The purpose of this study was to analyse the structure and contents of clothing construction field in the middle-school textbook of technology and home economics with curriculum revised in 2007. 10 kinds of textbooks certified by Ministry of Education, Science and Technology in 2010 were selected and home economics field in them was the subject of this study with the external characteristics such as textbook system, instruction area and contents included. This study was content analysis-oriented and descriptive with data calculated by frequency and percentage. The findings are as follows. In 'Reality of Home-Life', food life area showed highest as 34.4% followed by clothing life(28.7%) and dwelling life(28.7%). The portion of present teachers in writing staffs was highest while there was no textbook where professors of each field were included in writing staffs. The contents of costume society, clothes material, clothes management from section 'Choice and Management of Clothes' in the chapter 'Family Life' and clothing construction from the section 'Making and Modifying Clothes' of the chapter 'Reality of Home-Life' were analysed in this study. Clothing construction-related fields were suggested separately by method of measurement, construction theory, drafting theory, drafting an original form, fabric trimming, utilization of the original form, length measure, cloth cutting, hand stitching, how to use a sewing machine, basting and correction, needlework process, completion, and evaluation. The contents of the unit 'Making Shorts' were about clothing construction theory, human body measurement, drafting an original form, process of making shorts and activity/research/experience. The contents of the unit 'Wearing Clothes after Fixing' in clothing construction-related field were about how to fix and recycling.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37C no.9
• /
• pp.851-858
• /
• 2012

Java crypto library such as SUN JCA/JCE or BC JCE is generally used to implement secure applications for smart devices using Android platform. Programming functions written by Java language are launched and executed inside Java Virtual Machine (JVM), thereby difficult to use system hardware specific functionalities and degrading performance as well. In case of crypto primitive, few secure applications can use crypto primitive executing in JVM because both amount of computing and complexity of such primitives are very high. From the aspect of performance, in particular, time sensitive real time applications such as streaming services or secure application frequently applying public key based crypto algorithm cannot use Java crypto library. To solve the problem, we design and implement crypto library which employ JNI and NDK methods to directly access functions that implemented by native language such as C or C++. The proposed Java Crypto provider supports faster execution. Also developer can use our provider in the same way by writing traditional Java crypto library.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2005.05a
• /
• pp.1-9
• /
• 2005

Today, according to tremendous development of B2B e-commerce, B2B e-marketplaces which accomplish various types of transactions through a number of buyers and sellers on online are embossed importantly. However, buyers are unable to search correct products because of inconsistency of product information between buyers and sellers. This paper solved this problem as semantic web technology. Semantic Web is an extension of current Web in which information is given well-defined meaning, better enabling computers and people to work in cooperation. The Semantic Web aims at machine-processable information. Its underlying technologies are RDF, RDF Schema, and ontology as the shared formal conceptualization of particular domains. In this paper, we present an implementation of Semantic Web enabled search system for B2B E-Marketplace domains. The system exploits OWL as the standard ontology language proposed by W3C and the Jena which is a Semantic Web toolkit, namely a Java framework writing Semantic Web applications. Finally, we summarize our experiences and discuss future research topics.

• The KIPS Transactions:PartA
• /
• v.12A no.7 s.97
• /
• pp.563-570
• /
• 2005

Hardware and software codesign framework called PeaCE(Ptolemy extension as a Cod sign Environment) allows to express both data flow and control flow. To formally verify an fFSM specification which expresses control flow in PeaCE, the step semantics of the model was defined. In this paper, we introduce the automatic verification tool developed by formal semantics of previous work. This tool uses the SMV as inner model checker md, through our tool, users can formally verify some important bugs such as race condition, ambiguous transition, and circulartransition without directly writing logical formulae.