• Journal of KIISE:Computing Practices and Letters
• /
• 제15권12호
• /
• pp.988-992
• /
• 2009

In this paper, we attempted to extract binary relations between terminologies using composite kernels consisting of convolution parse tree kernels and WordNet verb synset vector kernels which explain the semantic relationships between two entities in a sentence. In order to evaluate the performance of our system, we used three domain specific test collections. The experimental results demonstrate the superiority of our system in all the targeted collection. Especially, the increase in the effectiveness on KREC 2008, 8% in F1, shows that the core contexts around the entities play an important role in boosting the entire performance of relation extraction.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 한국지능정보시스템학회 2002년도 추계정기학술대회
• /
• pp.328-335
• /
• 2002

최근 들어 인터넷 기술의 발달로 웹 상에 많은 자료들이 산재해 있습니다. 사용자가 원하는 정보를 검색하기 위해서 키워드 검색을 이용하고 있는데 이러한 키워드 검색은 사용자들이 입력한 단편적인 정보에 바탕하여 검색하고 검색된 결과들을 자체적인 기준으로 순위를 매겨 나열식으로 제시하고 있다. 이러한 경우 사용자들의 생각과는 다르게 결과가 제시될 수 있다. 따라서 사용자들의 검색 시간을 줄이고 편리하게 검색하기 위한 환경의 필요성이 높아지고 있다. 본 논문에서는 Suffix Tree 알고리즘을 사용하여 관련있는 문서들을 분류하고 각각의 분류된 클러스터에 제목을 생성하기 위하여 문서 빈도수, 단어 빈도수와 역문서 빈도수, 카이 검정, 공통 정보, 엔트로피 방법을 비교 평가하여 제목을 생성하는데 어떠한 방법이 가장 효과적인지 알아보기 위해 비교 평가해본 결과 문서빈도수가 TF-IDF보다 약 10%정도 성능이 좋은 결과를 보여주었다.

• Proceedings of the KSRS Conference
• /
• 대한원격탐사학회 2007년도 Proceedings of ISRS 2007
• /
• pp.247-249
• /
• 2007

Crown parameters are important roles in tree species identification, because the canopy is the aggregate of all the crowns. However, crown measurements with spaceborne image data have remained more difficult than on aerial photographs since trees show more structural detail at higher resolutions. This recognized problem led to the initiation of the research to determine if high resolution satellite image data could be used to identify and classify single tree species. In this paper, shape parameters derived from pixel-based crown area measurements and texture features derived from GLCM parameters in QuickBird image were tested and compared for individual tree species identification. As expected, initial studies have shown that the crown parameters and the canopy texture parameters provided a differentiating method between coniferous trees and broad-leaved trees within the compartment(less than forest stand) for single extraction from spaceborne high resolution image.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• 제32B권11호
• /
• pp.1536-1544
• /
• 1995

We propose a design scheme of a binary decision tree and apply it to the tire tread pattern recognition problem. In this scheme, a binary decision tree is constructed by using fuzzy C-means( FCM ) algorithm. All the available features are used while clustering. At each node, the best feature or feature subset among these available features is selected based on proposed similarity measure. The decision tree can be used for the classification of unknown patterns. The proposed design scheme is applied to the tire tread pattern recognition problem. The design procedure including feature extraction is described. Experimental results are given to show the usefulness of this scheme.

• Journal of Korea Spatial Information System Society
• /
• 제7권1호
• /
• pp.13-24
• /
• 2005

Objects, which change their positions over time such as cars, are called moving objects. Trajectories of a moving object have large volumes because trajectories are accumulated. Efficient indexing techniques for searching these large volumes of trajectories are needed in the moving object databases. Especially the TB-tree which supports bundling trajectories is suitable for processing combined queries which have 2 steps: first step is selecting trajectories (range search), next is selecting the parts of each trajectory (trajectory search). But the TB-tree has unnecessary disk accesses cause of lack of spatial discrimination in range queries. In this paper, we propose and implement the splitting polity which can reduce dead spaces of non-leaf node in order to process range queries efficiently. The policy has better performance about range queries than the TB-tree as well as the advantages of the TB-tree, such as highly space utilization and efficient trajectory extraction. This paper shows that the newly proposed split policy has better performance in processing the range queries than that of the TB-tree by experimental evaluation.

• Proceedings of the Korean Society of Plant Pathology Conference
• /
• 한국식물병리학회 2003년도 정기총회 및 추계학술발표회
• /
• pp.139.1-139
• /
• 2003

Tissues from woody plant contain higher amount of phenolic compounds and polysaccharides, which give inhibitory effects on reverse transcriptase and/or Taq ploymerase. The common multiple-step protocols using several additives to inhibit polyphenoic compounds during nucleic acid extraction are time consuming and laborious. Sodium sulfite (Na$_2$SO$_3$) was used as inhibitor of polyphenolic oxidases in extraction buffer and compare it's effect between commercial RNA extraction kit and small-scale double-stranded RNA (dsRNA) extraction by RT-PCR. During nucleic acid extraction procedure, addition of 0.5％-1.5％ (w/v) sodium sulfite to Iysis buffer or STE buffer resulted in lighter color change than extracts without sodium sulfite and improve the RT-PCR detection. When commercial RNA extraction kit used, optimal concentration of sodium sulfite were variable according to the host plant. However, using dsRNA as RT-PCR template, 1.5％ sodium sulfite in STE buffer improves the detection of both viruses and unspecific amplifications were reduced significantly, Furthermore, when viruses existed at low titers in host plant, small-scale dsRNA extractions were very reliable.

• Fashion & Textile Research Journal
• /
• 제8권4호
• /
• pp.465-470
• /
• 2006

Walnut hull is a by-product from the Walnut tree, used as natural dyestuff from ancient times. This study was done to examine the effects of extractive conditions on the properties of walnut hull extracts for making efficient use of the walnut hull as a natural colorant. Aqueous extracts of walnut hull were prepared at various extractive concentration, temperature and time. Then they were characterized using UV-Vis. Spectrophotometer, FT-IR Spectrometer, Prep Liquid Chromatography, and Energy dispersive X-ray spectrometer. The aqueous extracts have two absorbency peaks of UV-Vis. Spectrum, shoulder type peak in the range of 270-280 nm and broad type band around 420 nm. Intensity of absorbency is increased with increase of extraction concentration and time. However, Boiling temperature extraction method showed the most efficiency of all. Intensity of absorbency is also affected by extraction pH. The Prep LC examined two kinds of isolated colorant with different molecular weight. FT-IR spectra of hull extracts showed an absorption band around $3400cm^{-1}$, the peaks at $1700-1600cm^{-1}$, which are characteristic of aromatic compounds with unsaturated ketone and benzene ring. It showed that the extraction contained some mineral ions, such as K, Ca, Si, Mg.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제12권4호
• /
• pp.308-318
• /
• 2012

Captions include information which relates to the images. In order to obtain the information in the captions, text extraction methods from images have been developed. However, most existing methods can be applied to captions with a fixed height or stroke width using fixed pixel-size or block-size operators which are derived from morphological supposition. We propose an edge connected components based method that can extract Korean captions that are composed of various sizes and fonts. We analyze the properties of edge connected components embedding captions and build a decision tree which discriminates edge connected components which include captions from ones which do not. The images for the experiment are collected from broadcast programs such as documentaries and news programs which include captions with various heights and fonts. We evaluate our proposed method by comparing the performance of the latent caption area extraction. The experiment shows that the proposed method can efficiently extract various sizes of Korean captions.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제15권5호
• /
• pp.3213-3219
• /
• 2014

Currently, increase of greenhouse gas has had a signigicant impact on climate change in urbanization. As a result, the government has been looking for ways to take advantage of the trees that generate oxygen and reduce carbon dioxide for the prevention of climate change. It is essential to extract individual tree for calculating the amount of carbon dioxide reduction of trees. Aerial LIDAR data have three-dimensional information of building as well as trees as form of point clouds. In this study, automated model was developed to extract individual tree using aerial LIDAR data. For this purpose, we established a methodology for extracting trees and then proceeded the process of developing it as an automated model based on model builder of ArcGIS Software. In order to evaluate the applicability of the developed model, the model was compared with commercial software in study area located in Yongin City. Through the experimental result, the proposed model was extract trees 9.91% higher than commercial software. From this results, it was found that the model effectively extracted trees.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• 제23권6호
• /
• pp.711-718
• /
• 2019

BDD is a well-known alternative to the conventional Boolean logic method in fault tree analysis. As the size of fault tree increases, the calculation time and computer resources for BDD dramatically increase. A new failure path search and path restructure method is proposed for efficient calculation of CS and MCS from BDD. Failure path grouping and bottom-up path search is proved to be efficient in failure path search in BDD and path restructure is also proved to be used in order to reduce the number of CS comparisons for MCS extraction. With these newly proposed methods, the top event probability can be calculated using the probability by ASDMP(Approximate Sum of Disjoint MCS Products), which is shown to be equivalent to the result by the conventional MCUB(Minimal Cut Upper Bound) probability.