• Proceedings of the IEEK Conference
• /
• 2003.07d
• /
• pp.1311-1314
• /
• 2003

In this paper, we propose a method to measure document similarity. First, we have exploited single-term method that extracts nouns by using a lexical analyzer as a preprocessing step to match one index to one noun. In spite of irrelevance between documents, possibility of increasing document similarity is high with this method. For this reason, a term-phrase method has been reported. This method constructs co-occurrence between two words as an index to measure document similarity. In this paper, we tried another method that combine these two methods to compensate the problems in these two methods. Six types of features are extracted from two input documents, and they are fed into a neural network to calculate the final value of document similarity. Reliability of our method has been proved by an experiment of document retrieval.

• Annual Conference on Human and Language Technology
• /
• 1997.10a
• /
• pp.332-338
• /
• 1997

어휘 정보는 구문 구조의 중의성을 해결하는데 중요한 정보원으로서 작용할 수 있다. 본 논문에서는 입력 문장에 대한 구조적 중의성을 해결하는데 확률 구문 규칙뿐만 아니라, 어휘간에 발생할 수 있는 공기 정보를 사용할 수 있는 확률 모델을 제안한다. 제안된 확률 모델에 대하여 실험 데이타에 대해 평가한 결과 약 84%정도의 구문 분석 정확도를 얻을 수 있었다.

• 한국생태학회:학술대회논문집
• /
• 1999.05a
• /
• pp.55-62
• /
• 1999

Four species of Paraphysomonas collected from the fast- ice covered area Syowa Station, East Ongul Island ($69^{\circ}00'S,\;39^{\circ}35'$) ,Antarctica occurred in the seawater throughout the year and occasionally in the sea ice. P.. antarctica is distributed to a water depth of 35m at 51.3 during the period from August 1983 to January 1984 and also down to 600m St. 5 in September 1983 at cell concentrations of 300-350 cells/ml. The Paraphysomonas spp. were dominant during the period from July to November 1983 in the area studied. The mode of the occurrence and vertical distribution of Paraphysomonas apparently coresponds to those of the bacteria and orgarnic debris-like matter in the seawater. The main components of the plankton population in the area studied, under ice-covered conditions, are Paraphysomonas, Choanoflagellates and bacteria. This work clarified that Paraphysomonas is one o f the most important bacterivores in the microbial loop of the Antarctic marine ecosystem.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.7 no.6
• /
• pp.49-56
• /
• 1998

Super precision working technique and machine tool have been continually developed thanks to advanced electronic field. To obtain good result. it is necessary to investigate surface in grinding with $mu extrm{m}$ level. There were quite many researches to satisfy these demands by using non-contact methods through the computer vision. In this study, the texture of working surface was analyzed. co-occurrence matrices was obtained from the surface roughness. Texture parameter was obtained using position operator composed of $ heta$, d according to variation of angle direction and distance. As a result, it was found that surface texture was more affected by direction($\theta$) than distance(d).

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.9 no.2
• /
• pp.144-150
• /
• 2000

Surface roughness is one of the most important parameters to estimate quality of products. As this reason so many studies were car-ried out through various attempts that were contact or non-contact using computer vision. Even through these efforts there were few good results in this research., however texture analysis making a important role to solve these problems in various fields including universe aviation living thing and fibers. In this study feature value of co-occurrence matrix was calculated by statistic method and roughness value of worked surface was classified, of it. Experiment was carried out using input vector of neural network with characteristic value of texture calculated from worked surface image. It's found that recognition rate of 74% was obtained when adapting texture features. In order to enhance recogni-tion rate combination type in characteristics value of texture was changed into input vector. As a result high recognition rate of 92.6% was obtained through these processes.

• Macromolecular Research
• /
• v.9 no.1
• /
• pp.51-65
• /
• 2001

Two-stage emulsion polymerizations of hydrophobic monomers on hydrophilic seed polymer particles were carried out to make core-shell composite particles. It was found that the loci of polymerization in the second stage were the surface layer of the hydrophilic seed latex particles, and that it has resulted in the formation of either eccentric core-shell particles with the core exposed to the aqueous phase or aggregated nonspherical composite particles with the shell attached on the seed surface as many small separated particles. The driving force of these phenomena is related to the gain in free energy of the system in going from the hydrophobic polymer-water interface to hydrophilic polymer-water interface. Thermodynamic analysis of the present polymerization system, which was based on spreading coefficients, supported the likely occurrence of such nonspherical particles due to the combined effects of interfacial free energies and phase separation between the two polymer phases. A hypothetical pathway was proposed to prepare hydrophilic core-hydrophobic shell composite latex particles, which is based on the concept of opposing driving and resistance forces for the phase migration. It was found that the viscosity of the monomer-swollen polymer phase played important role in the formation of particle morphology.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2000.04a
• /
• pp.328-331
• /
• 2000

In this paper , we propose a texture feature extraction method of reduce the massive computational time on extracting texture, features of large sized medical such as MRI, CT-scan , and an index structure, called GLTFT, to speed up the retrieval performance. For these, the original image is transformed into a compressed image by Wavelet transform , and textural features such as contrast, energy, entropy, and homogeneity of the compressed image is extracted by using GLCM(Gray Level Co-occurrence Metrix) . The proposed index structure is organized by using the textural features. The processing in compressed domain can give the solution of storage space and the reduction of computational time of feature extracting . And , by GLTFT index structure, image retrieval performance can be expected to be improved by reducing the retrieval range . Our experiment on 270 MRIs as image database shows that shows that such expectation can be got.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.12a
• /
• pp.117-120
• /
• 2001

In this work, we propose a new method of extracting and weighting representative keywords(RKs) from a few documents that might interest a user. In order to extract RKs, we first extract candidate terms and then choose a number of terms called initial representative keywords (IRKS) from them through fuzzy inference. Then, by expanding and reweighting IRKS using term co-occurrence similarity, the final RKs are obtained. Performance of our approach is heavily influenced by effectiveness of selection method of IRKS so that we choose fuzzy inference because it is more effective in handling the uncertainty inherent in selecting representative keywords of documents. The problem addressed in this paper can be viewed as the one of calculating center of document vectors. So, to show the usefulness of our approach, we compare with two famous methods - Rocchio and Widrow-Hoff - on a number of documents collections. The results show that our approach outperforms the other approaches.

• Journal of the Korea Computer Industry Society
• /
• v.4 no.10
• /
• pp.587-598
• /
• 2003

This paper construct two-level word similarity relation matrices about title and to scientific treatise. As guide keyword similarity relation matrices which is constructed to co-occurrence frequency base same time keeps recall rater by query expansion by tolerance relation, it is index structure to improve the precision rate by two-level contents base retrieval. Therefore, draw area knowledge through subject analysis and reasoned user's information request and area knowledge to fuzzy logic base. This research is research to improve vocabulary mismatch problem and information expression having essentially on query.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.201-204
• /
• 2016

본 논문에서는 문서의 공기관계를 통해 추출된 문서의 특징을 이용하여 유사 보고서를 판별하는 시스템을 제안한다. 국가 R&D 보고서의 XML형식 파일에서 텍스트를 추출 후, 문장 단위로 나누어 각 문장의 공기관계를 추출한다. 그 후 공기관계의 노드와 엣지를 문서에 추가하고, 노드로 사용된 단어만 남기고 나머지 단어는 제외한다. 그리고 이것을 문서의 특징으로 삼고 유사도 계산을 한다. 이 때, 유사도 계산은 코사인 유사도를 사용한다. 실험결과, 국가 R&D문서 유사도 계산에서 제안된 방법이 기존의 방법보다 높은 분류율을 보여주었다.