• 한국멀티미디어학회논문지
• /
• 제17권2호
• /
• pp.232-239
• /
• 2014

온라인에서의 글쓰기가 늘어나면서, 기계학습을 통해 이를 분류하는 연구가 늘고 있다. 그럼에도 불구하고 한국어로 작성된 마이크로블로그를 대상으로 한 연구는 많지 않다. 또한 통계적으로 기계학습을 평가한 연구를 찾아보기 힘들다. 본 논문에서는 트위터를 대상으로, 표본을 추출하고, 형태소와 음절을 자질로 사용하여 기계학습에 따라 감정을 분류하였다. 그 결과 약 76%정도 트위터에 포함된 감정이 분류되었다. Support Vector Machine이 Na$\ddot{i}$ve Bayes보다 정확했고, 선형모델도 비구조적인 텍스트 처리에 비선형모델에 상응하는 정확성을 보였다. 또한 형태소가 음절 자질에 비해 높은 정확성을 보이지 않았다.

• 전기학회논문지
• /
• 제66권8호
• /
• pp.1272-1277
• /
• 2017

To assist radiologists for the characterization of breast masses, Computer-aided Diagnosis(CADx) system has been studied. The CADx system can improve the diagnostic accuracy of radiologists by providing objective information about breast masses. Morphological and texture features were extracted from the breast ultrasound images. Based on extracted features, the CADx system retrieves masses that are similar to a query mass from a reference library using a k-nearest neighbor (k-NN) approach. Eight similarity measures of distance, Euclidean, Chebyshev(Minkowski family), Canberra, Lorentzian($F_2$ family), Wave Hedges, Motyka(Intersection family), and Cosine, Dice(Inner Product family) are evaluated by ROC(Receiver Operating Characteristic) analysis. The Inner Product family measure used with the k-NN classifier provided slightly higher performance for classification of malignant and benign masses than those with the Minkowski, $F_2$, and Intersection family measures.

• BMB Reports
• /
• 제31권2호
• /
• pp.183-188
• /
• 1998

2-chloroethylethyl sulfide (CEES) is an alkylating agent that readily reacts with a wide variety of biological molecules causing metabolic abnormality. The mechanism of cell death during CEES injury is poorly understood. We have examined the effect of exposure of thymocytes with various concentrations of CEES to determine the pattern of cell death in thymocytes injury induced by CEES. In the present study, we show that two patterns of cell death occurred by either one of two mechanisms: apoptosis and necrosis. Exposure to low level of CEES (100 ${\mu}M$) for 5 h caused an induction of apoptosis on thymocytes, as identified by the following criteria: DNA fragmentation visualized by the characteristic "ladder" pattern was observed upon agarose gel electrophoresis and morphological features were revealed by microscopical observations. In contrast, exposure to high levels of CEES (500 ${\mu}M$) induce necrotic features such as cell lysis. Thus, depending on the concentrations, CEES can result in either apoptotic or necrotic cell damage. Our findings suggest that thymocytes which are not killed directly, but merely injured by low levels of CEES, are able to activate an internally-programmed cell death mechanism, whereas thymocytes receiving severe damages apparently can not.

• 동의생리병리학회지
• /
• 제16권3호
• /
• pp.602-608
• /
• 2002

To investigate whether Puerariae radix (PR) possesses protective effect against ethanol (EtOH)-inducecl apoptosis in the central nervous system, MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay, flow cytometric analysis, DNA fragmentation assay. and reverse transcription-polymerase chain reaction (RT-PCR) were performed on human neuroblastoma cell line SK-N-MC. Morphological and biochemical analyses demonstrated that SK-N-MC cells treated with EtOH exhibit classical apoptotic features. On the other hand, cells pre-treated with PR prior to EtOH exposure showed decreased occurrence of classical apoptotic features. In addition, it was shown that PR pre-treatment inhibits EtOH-induced increases in the levels of mRNA expression of bax and caspase-3, while it further enhances the level of bcl-2 expression. These results suggest that PR may exert protective effects against EtOH-induced apoptosis in human neuroblastoma cells.

• 한국산업정보학회논문지
• /
• 제11권1호
• /
• pp.13-20
• /
• 2006

본 연구에서는 어휘정보와 개념정보를 기반으로 스팸메일 필터링 시스템을 구축하였다. 스팸메일을 판별할 수 있는 정보를 두 가지로 구분하였는데, 확실한 정보군은 송신자 정보, URL, 그리고 최근 스팸 키워드 리스트이며, 덜 확실한 정보군은 메일 본문에서 추출한 단어목록과 개념코드이다. 먼저 확실한 정보군을 이용하여 스팸메일을 분류하고 그다음 덜 확실한 정보군을 이용하였다. 메일 본문에 포함된 어휘정보와 개념코드는 SVM 기계학습을 한 후 사용된다. 본 연구의 결과, 더 많은 어휘정보를 특징벡터로 사용하였을 때 스팸 정확률이 상승하였으며 추가로 개념코드를 특징벡터에 포함시켰을 때 스팸 재현율이 상승하였다.

• 환경생물
• /
• 제36권2호
• /
• pp.150-155
• /
• 2018

Two siphonous green algae were collected from the eastern coast of Korea. These species share the typical features of Derbesia sporophytes, such as erect and prostrate siphonous filaments and the presence of basal septum in lateral branches. One is characterized by the combined features of a relatively small tufted sporophytic thalli arising from a rhizoidal base, subdichotomously branched filaments with a basal septum and one pyrenoid per subspherical to lenticular chloroplast. The other shows a larger sporophytic thallus, sparsely subdichotomously branched filaments with a basal septum and one pyrenoid per spherical chloroplast. In a phylogenetic tree based on rbcL sequence, the two above-mentioned Korean algae nest in the same clades as Derbesia minima and D. indica, respectively. The genetic distance between the sequences within the clades was 0.5-0.8%, which is considered to be included in the intra-specific range for the genus. These two siphonous Korean algae are identified as D. minima and D. indica, respectively, based on the morphological and molecular analyses. These species are newly recorded in the Korean marine algal flora herein.

• 한국컴퓨터정보학회논문지
• /
• 제25권8호
• /
• pp.65-72
• /
• 2020

금융, 통신 등의 응용 분야에서 빅데이터는 광범위하게 활용되고 있으며, 빅데이터 분석 시장은 해마다 크게 성장하고 있다. 이에 반해 소상공인들의 빅데이터 활용 실적은 저조하며, 이는 기존 시스템이 소상공인들의 여건을 충분히 반영하지 못하는 동시에 서비스 이용 가격 역시 높다는 점에 기인한다. 이를 해결하기 위한 노력의 일환으로, 본 논문에서는 소상공인에 특화된 빅데이터 분석 서비스를 제공하는 새로운 플랫폼을 개발, 제안한다. 먼저 소셜 빅데이터 분석과 관련한 기존 연구들을 비교하고, 소상공인의 마케팅을 돕기 위해 필요한 서비스 지표들을 추출한다. 다음으로 도출된 지표들을 구현한 프로토타입 시스템을 소개하고, 구현을 통해 얻어진 시스템 완성에 필요한 기술적인 이슈들을 논의한다.

• ETRI Journal
• /
• 제38권1호
• /
• pp.174-184
• /
• 2016

Information identification with image data by means of low-level visual features has evolved as a challenging research domain. Conventional text-based mapping of image data has been gradually replaced by content-based techniques of image identification. Feature extraction from image content plays a crucial role in facilitating content-based detection processes. In this paper, the authors have proposed four different techniques for multiview feature extraction from images. The efficiency of extracted feature vectors for content-based image classification and retrieval is evaluated by means of fusion-based and data standardization-based techniques. It is observed that the latter surpasses the former. The proposed methods outclass state-of-the-art techniques for content-based image identification and show an average increase in precision of 17.71% and 22.78% for classification and retrieval, respectively. Three public datasets - Wang; Oliva and Torralba (OT-Scene); and Corel - are used for verification purposes. The research findings are statistically validated by conducting a paired t-test.

• 건축역사연구
• /
• 제24권1호
• /
• pp.41-50
• /
• 2015

This study aims to investigate the physical characteristics and architectural ornaments of the walls (DamJang) and their bordering area that defines the Rear Garden of Changdeok-gung, one of the Royal Palaces in Seoul. This area, centered on Buyong-ji(부용지) and Aeryeon-ji(애련지), features the morphological diversity of buildings, walls, gates and stone bases in the palace. The findings are summarized as follows: First, DamJang, as a basic architectural element for the space organization, takes a set of various forms featuring different construction materials while responding positively to the physiographic nature of the surroundings; Second, DamJang along with their gates, also features different types of ornamental expressions which also suggest the hierarchy of its building and space; Third, typical of the traditional garden design in Korea are a group of DamJang standing as independent structures mostly around Buyong-ji and Aeryeon-ji whose finishing conditions also differ based on the nature of the space; Fourth, among commonly observed examples of DamJang in the Rear Garden and Changdeok-gung palace-wide, is a 'T' shaped wall whose heights and materials function as a design element for the garden.

• 대한의용생체공학회:의공학회지
• /
• 제29권3호
• /
• pp.173-178
• /
• 2008

Recent advancements in the electrospinning method enable the production of ultrafine solid and continuous fibers with diameters ranging from a few nanometers to a few hundred nanometers with controlled surface and morphological features. A wide range of biopolymers can be electrospun into mats with a specific fiber arrangement and structural integrity. These features of nanofiber mats are morphologically similar to the extracellular matrix of natural tissues, which are characterized by a wide pore diameter distribution, a high porosity, effective mechanical properties, and specific biochemical properties. This has resulted in various kinds of applications for polymer nanofibers in the field of biomedicine and biotechnology. The current emphasis of research is on exploiting these properties and focusing on determining the appropriate conditions for electrospinning various biopolymers for biomedical applications, including scaffolds used in tissue engineering, wound dressing, drug delivery, artificial organs, and vascular grafts, and for protective shields in specialty fabrics. This paper reviews the research on biomedical applications of electrospun nanofibers.