• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06a
• /
• pp.79-81
• /
• 2006

커널(Kernel)을 이용한 분류 방법은 넓은 마진(large margin) 분류기로서 SVM(Support Vector Machine)을 주로 사용하게 된다 하지만, 이 방법은 라그랑제 파라미터(Lagrange Parameter)의 최적화 과정을 포함함으로써 학습 과정을 쉽지 않게 만든다. 이 최적화 과정은 특히 DNA computing과 같은 단순한 과정의 설계를 통해 결과를 얻어야 하는 새로운 계산 모델에 커널을 적용하고자 했을 경우 큰 장벽이 된다. 본 논문에서는 넓은 마진을 목표로 하는 최적화 과정이 아닌 다른 라벨(label)의 데이터간의 경계 파악을 위한 간단한 커널 갱신 방법의 도입을 통해 분류기를 설계한다. 이 방법을 가우시안 커널에 적용시켜 본 결과, 반복을 통해 데이터의 구조를 찾아갈 수 있는 특성을 보여주며, 결국 넓은 마진의 최적화된 파라미터를 찾게 됨을 보여준다. 본 논문에서는 이 최적화 방법을 DNA 분자를 이용한 커널 생성 모델인 DNA 커널에 적용시켰을 때 잘 알려진 AML/ALL 데이터를 잘 분류해 냄을 보여준다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.259-261
• /
• 2005

분류 방법으로서의 SVM(Support Vector Machine)은 커널 방법과 함께 사용됨으로써 그 유용성을 크게 향상시켰다. 커널 방법은 일반적으로 입력 데이터의 자질(feature)로 나타내는 공간으로부터 높은 차원의 공간으로 데이터를 사상(mapping)시키는 역할을 하게 되나, 기본적으로는 데이터간에 새로운 거리(metric)를 부설해주는 역할을 하는 것이다. 지금까지 나온 다양한 커널 방법은 구조화된(structured) 데이터에 대해 커널 형태로 거리를 부여하는 방법을 제시한다. 본 논문에서는 DNA의 작용을 모델링하여 만든 새로운 커널이 miRNA(micro RNA)와 mRNA(messenger RNA)쌍에 대한 발현 여부를 분류해 내기 위해 커널 형식으로 거리를 부여하는 방법을 보인다. 이 방법은 실리콘 컴퓨터가 아닌 실제 DNA분자로 실험할 수 있도록 설계된 것을 고려할 때 여러 종류의 DNA 코드를 분석하는 데 사용될 수 있는 새로운 분자컴퓨팅 방법이다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.2
• /
• pp.49-59
• /
• 2007

This paper is about an system which can perform automatic semantic annotation to actualize "Semantic Web." Since it is impossible to tag numerous documents manually in the web, it is necessary to gather large Korean web documents as training data, and extract features by using natural language techniques and a thesaurus. After doing these, we constructed concept classifiers through the SVM (support vector machine) teaming algorithm. According to the characteristics of Korean language, morphological analysis and syntax analysis were used in this system to extract feature information. Based on these analyses, the concept code is mapped with Kadokawa thesaurus, which made it possible to map similar words and phrase to one concept code, to make training vectors. This contributed to rise the recall of our system. Results of the experiment show the system has a some possibility of semantic annotation.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.118-125
• /
• 2020

This paper is conducted for automatically recommending and providing information services desired by users on websites of local governments and public institutions with vast amounts of information, In this system, we defined a method of collecting data based on the SiiRU CMS system that collects and preprocesses data, and a study that provides curation services (contents and menus) to users through a collaborative filtering algorithm based on machine learning. Also, the data used in the paper is conducted based on about 1 million data collected in 2019. The analyzed data can provide important information that cannot be easily accessed by providing a cloud tag service or recommended menu for users to conveniently view, and the environment configuration that can realize this service to local governments and public institutions is also provided.

• Journal of the Korea Society for Simulation
• /
• v.32 no.1
• /
• pp.13-21
• /
• 2023

This paper proposes a method of automatically generating a model through a machine learning technique by setting a hypothetical model in the form of a gray box or black box with unknown parameters, when the big data of the actual system is given. We implements the proposed framework and conducts experiments to find an appropriate model among various hypothesis models and compares the cost and fitness of them. As a result we find that the proposed framework works well with continuous systems that could be modeled with ordinary differential equation. This technique is expected to be used well for the purpose of automatically updating the consistency of the digital twin model or predicting the output for new inputs using recently generated big data.

• Journal of Information Technology and Architecture
• /
• v.9 no.4
• /
• pp.413-422
• /
• 2012

Many researchers have been studied to reveal human natural language to let machine understand its meaning by text based, page rank based or more. Particularly, it has been considered that URL and HTML Tag information in web documents are attracting people' attention again to analyze huge amount of web document automatically. In this paper, we propose a STW (Semantic Term Weight) approach based on syntactic and linguistic structure of web documents in order to classify what genres are. For the evaluation, we analyzed more than 1,000 documents from 20-Genre-collection corpus for training the documents based on SVM algorithm. Afterwards, we tested KI-04 corpus to evaluate performance of our proposed method. This paper measured their accuracy by classifying them into an experiment using STW and one without u sing STW. As the results, the proposed STW based approach showed approximately 10.2% which Is higher than one without use of STW.

• Restorative Dentistry and Endodontics
• /
• v.32 no.3
• /
• pp.236-247
• /
• 2007

The purpose of this study was to evaluate the effects of cyanate methacylate on the shear bond strengths to bovine dentin surfaces as a dentin primers. Seven experimental adhesives were made with different mass fraction of Isocyanatoetylme-thacrylate (IEM), 40wt% HEMA (Wako Pure Chemical Industries Osaka, Japan), 0.6% camphoroquinone, 0.4% amine and ethanol as balance dentin bonding agents (0, 2, 4, 6, 8, 10, 12%) were made and applied on the surface of bovine dentin specimens of 7 experimental groups. Shear bond strengths were measured using a universal testing machine (Instro 4466). To identify the ratio and modes of cohesive failures, microscopic examinationn was performed. The ultra-structure of resin tags were observed under scanning electron microscope. The results were as follows ; 1) A higher shear bond strengths (33.62 MPa) in group 8% of Cyanate methacrylate to dentin were found, but there were no statistically significancy between Groups (p > 0.05). 2) The higher ratio of cohesive failures mode in group 2, 6, an 10% could be seen than that in any other groups. 3) A shorter resin tags were observed in all experimental groups. This could be resulted that the preventing from the cyanate methacrylate penetrate into dentin owing to reacting it with dentin collagen. Therefore the resin tags were shorter in lengths. Whether the higher bonding strengths of dentin bonding agents can be affected was not been assured with statistic results. The results indicated that the relation between tensile strengths of the dentin adhesives to bovine dentin and resin tags formed into the dentin could not affected. The main reason of increasing the shear bond strength to bovine dentin in experimental groups could not be assured.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.32 no.4
• /
• pp.262-272
• /
• 2020

A large number of studies on wave breaking have been carried out, and many experimental data have been documented. Moreover, on the basis of various experimental data set, many empirical or semi-empirical formulas based primarily on regression analysis have been proposed to quantitatively estimate wave breaking for engineering applications. However, wave breaking has an inherent variability, which imply that a linear statistical approach such as linear regression analysis might be inadequate. This study presents an alternative nonlinear method using an neural network, one of the machine learning methods, to estimate breaking wave height and breaking depth. The neural network is modeled using Tensorflow, a machine learning open source platform distributed by Google. The neural network is trained by randomly selecting the collected experimental data, and the trained neural network is evaluated using data not used for learning process. The results for wave breaking height and depth predicted by fully trained neural network are more accurate than those obtained by existing empirical formulas. These results show that neural network is an useful tool for the prediction of wave breaking.

• The Journal of the Korea Contents Association
• /
• v.9 no.11
• /
• pp.74-79
• /
• 2009

This paper proposes a hardware structure and associated finite state machine designs sharing key scheduling circuitry to enhance the performance of the block cypher algorithm, HIGHT. It also introduces an efficient protocol applicable to RFID systems comprising the HIGHT block cipher algorithm. The new HIGHT structure occupies an area size small enough to accommodate tag applications. The structure yields twice higher performance them conventional HIGHT algorithms. The proposed protocol overcomes the security vulnerability of RFID tags and thereby strengthens the security of personal information.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.1
• /
• pp.125-137
• /
• 2009

Smart card has a small sized micro computer chip. This chip contains processor, RAM, ROM, clock, bus system and crypto-co-processor. Hence it is more expensive, complicated and secure chip compared with RFID tag. The main application area of smart card is e-banking and secure communications. There are two kinds of smart card platforms; open platform and closed one. Java card is the most popular open platform because of its security, platform independency, fast developing cycle. However, the speed of Java card is slower than other ones, hence there have been hot research topics to improve the performance of Java card. In this paper, we propose an efficient transaction buffer management to improve the performance of Java card. The experimental result shows the advantage of our method.