범주별 대표 키워드에 의한 텍스트 분류

조태호 한국전문가시스템학회 1998 학술대회 Vol.2 No.1

A scheme of automatic document classification will be presented in this paper. So far, documents have been classified according to its contents manually. Therefore, it costs very much to assign a category to them because human investigates their contents. As the amount of data stored in storage media is increased exponentially it becomes necessary to store documents according to their category, to access them easily. It requires automatic text classification to store documents like that. The scheme for automatic text classification proposed in this paper, is based on the document indexing, in which a document is represented as a list of keywords. The number of common keywords between keywords from the document itself and representative keywords from a back data classifies documents. As an example, the proposed scheme is applied to Classify news articles into 3 categories; politics, sports, and business. The measurements of performance evaluation are classification rate, correctness rate, and classified correctness rate.

Table based Matching Algorithm for Soft Categorization of News Articles in Reuter 21578

조태호 한국멀티미디어학회 2008 멀티미디어학회논문지 Vol.11 No.6

This research proposes an alternative approach to machine learning based ones for text categorization. For using machine learning based approaches for any task of text mining, documents should be encoded into numerical vectors; it causes two problems: huge dimensionality and sparse distribution. Although there are various tasks of text mining such as text categorization, text clustering, and text summarization, the scope of this research is restricted to text categorization. The idea of this research is to avoid the two problems by encoding a document or documents into a table, instead of numerical vectors. Therefore, the goal of this research is to improve the performance of text categorization by proposing approaches, which are free from the two problems.

Single Pass Algorithm for Text Clustering by Encoding Documents into Tables

조태호 한국멀티미디어학회 2008 멀티미디어학회논문지 Vol.11 No.12

This research proposes a modified version of single pass algorithm specialized for text clustering. Encoding documents into numerical vectors for using the traditional version of single pass algorithm causes the two main problems: huge dimensionality and sparse distribution. Therefore, in order to address the two problems, this research modifies the single pass algorithm into its version where documents are encoded into not numerical vectors but other forms. In the proposed version, documents are mapped into tables and the operation on two tables is defined for using the single pass algorithm. The goal of this research is to improve the performance of single pass algorithm for text clustering by modifying it into the specialized version.

Modified Version of SVM for Text Categorization

조태호 한국지능시스템학회 2008 INTERNATIONAL JOURNAL of FUZZY LOGIC and INTELLIGE Vol.8 No.1

This research proposes a new strategy where documents are encoded into string vectors for text categorization and modified versions of SVM to be adaptable to string vectors. Traditionally, when the traditional version of SVM is used for pattern classification, raw data should be encoded into numerical vectors. This encoding may be difficult, depending on a given application area of pattern classification. For example, in text categorization, encoding full texts given as raw data into numerical vectors leads to two main problems: huge dimensionality and sparse distribution. In this research, we encode full texts into string vectors, and apply the modified version of SVM adaptable to string vectors for text categorization.

제57차 춘계학술대회 : 일반포스터 ; Minoxidil 치료로 호전된 양털모양 모발(Woolly Hair) 1예

조태호,최천필,이무형,김낙인,허충림,심우영 대한피부과학회 2005 초록집 Vol.43 No.10

Homepage상에서 사용되는 색채 선호도 연구

조태호 장안대학교 2007 長安論叢 Vol.27 No.-

오늘날에 있어서 computer의 활용은 사회 전반에 걸쳐 과히 혁명이라 하지 않을 수 없을 정도로 다방면에서 쓰이고 있다. 그 중에서도 인터넷은 급속히 발전하여 각종 정보를 공유하는 가장 중요한 수단이 되었다. 그러한 환경 하에서 현대에는 국제적으로 인적, 물적 교류가 증가하고 있고, 영상을 포함한 멀티미디어 통신이 중요한 역할을 담당하는 시대에 있어서 세계적인 시야로 인터넷에서 사용되는 색채의 인식에 대한 조사가 필요하다 하겠다. 특히 인간은 사물을 대할 때 시각을 적극 활용하며 색채 요소는 그 시각 반응의 대부분을 차지하는 중요한 요소라 할 수 있다. 따라서 웹디자인에서 색상의 중요성은 기업의 이미지를 홈페이지를 찾는 방문자들에게 강력하게 인지시킬 수 있는 가장 효과적인 디자인 요소일 것이다. 이에 본 연구는 국제적인 도메인인 .com을 사용하는 Homepage 자료를 다수 수집하여 세계인들의 색채 기호도와 인식도를 분석함으로써 웹디자인에서 색상의 적용을 어떻게 처리하는 것이 효율적인가 하는 것을 제시하는데 목적이 있다, 객관적인 분석을 위해서 www.yahoo.com을 통하여 총2477건의 Homepage 자료를 무작위로 수집하였고 색상, 명도, 배합색 등으로 분류하여 비교 분석하였다. 분석결과 배경색으로는 White(51.8%)를 가장 선호하며, 배합색으로는 White와 Black(l8.1%)을 같이 사용하는 경우가 가장 많았으며, 그 다음으로는 Blue 계열(l7.3%)의 색채를 많이 사용하는 것으로 나타났다.

智異山 노고단 일대의 파리類 및 季節的인 發生消長

조태호 晋州敎育大學校 科學敎育硏究所 2001 科學敎育硏究 Vol.27 No.-

This paper aims to explore the fauna of the fly, their seasonal prevalence and monthly changes in species diversity, along with altitudinal distribution of the predominant species of flies in the Nogodan area of Mt. Jiri. S. Korea. The conservation states of the natural environment of Mt. Jiri, between the years of 1995 and 2000, were compared according to fly species. The fly samples utilized during this research were collected monthly - from February to December - from three different fly traps during the years of 1995 and 2000. Trap #l was set at sea level (680m). Trap # 2 at 1,100m, and Trap #3 at 1,450m. Our results reveal that the fauna of flies consists of thirty-seven species which are categorized into five different families: a) twelve species of Calliphoridae; b) fifteen of Musidae: c) eight of Sarcophagidae: d) one of Scatophagidae and: e) one of Dryomyzidae. The common species included eighteen belonging to three different family types in two years. The index of species diversity (H') increased according to the altitude height of the surveyed sites in the months of July and August. With respect to seasonal prevalence, it was found that the number of female flies increases mainly during the months of May to June, while males increase mainly in June, indicating that May and June are flies' most prolific season. The order of the five dominant species researched during the years of 1995 and 2000, Hydrotaea dentipes, Chrysomyia pinguis, Calliphora lata, Fannia scalaris, and Aldrichina grahami, turned out to be the same but the composition rates for each speices varied. The patterns of seasonal prevalence of the five dominant species can be classified into three groups. H. dentipes. C. lata, and F. scalaris demonstrated great increases during the months of May and June, but sharp decreases during July and August; C. pinguis increased greatly in the months of July and August despite the summer season, A. grahami displayed steady increases in July and August. As for the relative abundance of the dominant species, H. dentipes had a great number of increases in the months of May and June, C. pinguis during August, C. lata in the months of April and October, but F. scalans and A. grahami generally kept low percentages (below 00%) during all seasons. In relation to the altitudinal distribution of the dominant species, H. dentipes. C. lata, F. scalaris, and A. grahami were mainly concentrated in T #l (680m) region in spring and fall, and in T #2 (1,100m) and T #3 (1,450m) regions in summer. However, it was found that most of C. pinguis' prefer to live in high air temperature regions. Despite some natural environmental changes causing the index in trap #l during the year 2000 to fluctuate at 71.1∼73.296, in general, all traps indicated high similarity index factors ranging from 83.3% to 87.0%.

자궁외임신 환자에 있어 다글라스와 혈액 β-hCG의 의의

조태호,이임순,이민철,고영익,최재호,양정분 대한산부인과학회 1990 Obstetrics & Gynecology Science Vol.33 No.6

1988년 4월 1일부터 1989년 5월 31일까지 순천향대학병원 산부인과에 내원한 자궁외임신 환자 29명을 대상으로 수술전 말초혈액과 수술시 복강내에 고여있던 혈액에서 β-hCG를 측정하여 파열형과 비파열형으로 나누어 비교분석하여 다음과 같은 결과를 얻었다. 1. 자궁외임신의 위치와 난관양상을 보면 연구대상 29예중 부위별로는 난관팽대부가 22예(76%), 체부가 5예(17.2%), 협부와 간질부는 각각 1예(3.4%)로 팽대부가 많았고, 난관양상은 파열형이 14예(48.2%), 비파열형 15예(51.8%)였다. 2. 혈청과 다글라스와 혈액에서의 β-hCG평균치를 비교시 혈청 2220.72 ± 661.888mIU/ml, 다글라스와 혈액 7338 ± 1331.83mIU/ml로 다글라스와 혈액에서 의미있게 높았고(P$lt;0.0001), 두 군에서 각각 파열형과 비파열형에 따른 β-hCG의 비교시 혈청에서는 파열형 3351.79 ± 1249.83mIU/ml, 비파열형 1165.73 ± 418.99mIU/ml, 다글라스와에서는 파열형 8499.29 ± 1694.30 mIU/ml, 비파열형 7155.73 ± 2067.11mIU/ml로 통계학적 의의는 없었다(P$gt;0.05). 3. 혈청 β-hCG치에 대한 다글라스와 혈액 β-hCG치의 비율의 범위는 최소 0.42에서 최고 85.93으로 평균은 13.35였고 혈청치가 낮을 수록 비율이 높아짐을 보이며, 난관 양상에 따라 비교하면 각각 9.78과 16.67로 비파열형에서 비율이 높았다. 4. 혈청과 다글라스와 혈액 β-hCG치의 상관관계는 파열형에서는 Y=1.05X + 4971.82(r=0.77, P$lt;0.05, N=14)로 상관관계가 높고 통계학적 의의도 있으나 비파열형에서는 Y=2.26X +4519.87(r=0.46, P=0.09, N=15)로 통계학적 의의는 없었다. 종합하면 Y=1.16X +5000.45(r=0.57, P$lt;0.05, N=29)로 통계학적 의의가 있었다. 이상에서 다글라스와 β-hCG를 측정하여 말초혈액의 β-hCG의 측정치와 비교함으로써 자궁외임신의 조기진단에 도움을 줄것으로 사료된다. The purpose of this study was to know whether the cul-de-sac blood β-hCG value serve as a tool for early diagnosis of ectopic pregnancy. Retrospective studies were made on serum β-hCG and cul-de-sac blood β-hCG values of 29 ectopic pregnancies, which were diagnosed by biopsy of surgical specimen, during 14 months from April 1, 1988 to May 31, 1989. The results were obtained as follows: 1. The mean value of β-hCG in ruptured ectopic pregnancy was 3351.79 ± 1249.81 mIU/ml (mean ± S.E.M.) in serum and 8499.29 ± 1694.30 mIU/ml in cul-de-sac blood. The mean value of β-hCG in unruptured group was 1165.73 ± 418.99 mIU/ml in serum and 7155.73 ± 2067.11 mIU/ml in cul-de-sac. When comparing the β-hCG level in serum of the ruptured group with that of the unruptured group, there was so significant value. Also when comparing the β-hCG level in cul-de-sac of the ruptured group with that of the unruptured group, there was no significant value. However, the β-hCG level in cul-de-sac (7338.83 ± 1331.83 mIU/ml) was significantly higher than serum β-hCG (2220.72 ± 661.88 mIU/ml). (P $lt; 0.0001). 2. The ratio of cul-de-sac to serum β-hCG had ranged from 0.42 to 85.93 and the mean ratio was 13.35. In according to decrease serum β-hCG level, the ratio was increased. The ratio was higher in unruptured group (mean : 16.67) than ruptured (9.78). 3. The correlation between serum and cul-de-sac β-hCG significant (r= 0.77, P $lt; 0.05) in ruptured group, and not significant (r= 0.46, P= 0.09) in unruptured. In all ectopic pregnancies, the correlation was significant (r= 0.57, P $lt; 0.05).

제26, 27회 세계 체조 선수권대회 경기 내용 연구 분석 : 철봉 종목을 중심으로

조태호,전영석 경희대학교 체육과학연구소 1995 體育學論文集 Vol.23 No.-

The study was made to research and analyze the contents of the exercise on the bar in the 26th and 27th world championship contest of gymnastics using films which recored 5 man each totaling 10, Ll. C(CHN), BLUCHNER. R(GER), SCHRBO. V(URS), KESWICK(U. S. A), SUPOLA. 2(HUN) in the 26th and CHARKOV.S(RUS), GHERMAN. M(ROM), SUPOLA(HUN), SHERBO. V(BRL), and CHARIPOV. R(UKR) in the 27th who participated in the final contest, about Difficulty, Composition and Exercise. The finding were as follows; 1)The difficulty of Exercise ; The number of Difficult "D" was 15 and "E" was 6 In the 27th contest against difficult "D" was 13 in the 26th contest. 2)The practical exercise ; The beginning of exercise was conducted at the point of increase of difficult value in the 27th contest and middle of exercise was also continue of difficulty, thus, vague increased and additional points of exercise was conducted in the same degree in the 26th and the 27th contest. 3)Special requirements ; There was no player who violated the special requirement in the 26th and 27th contest. 4)Additional points ; The additional points in the 26th contest was 0.60P and average point was 0.40P. In the 27th contest was 1.00P and average point was 0.72P. 5)General Results. As a result of review all the result, according to the requirement of newly revised score rule, the perfection and the connection of difficulty and difficulty, the practical quality of exercise has been conducted dynamically. Therefore, according to world-wide trend, with the more perfect exercise and new technics and connection of difficulties, we have to deal with developing additional value points techniques.

부인과 악성종양의 진단에 있어서 세침천자세포검사의 유용성

조태호,차상헌,문원실,이치훈,홍기언,허승재 대한산부인과학회 1992 Obstetrics & Gynecology Science Vol.35 No.9

저자들은 부인과 악성종양이 의심되는 46명의 환자에서 48예의 세침천자를 시행하여 다음과 같은 결과를 얻었다. 1. 진단결과 세침천자상 악성세포 양성인 진양성은 22예, 진음성은 19예, 위음성은 4예, 의중 1예, 부적합검체 4예였다. 2. 통계학적 결과 민감도는 84.6%이고 특이도는 100%, 위음성율 15.3% 이었으며 음성예측도는 82.8%이었다. To evaluate the efficacy of fine-needle aspiration in gynecologic oncology, 48 cases of fine-needle aspiration were performed on 46 patients suspected of having primary or recurrent gyneclolgic malignancy. There were 22 positive cases for malignant cell, 22 negative cases, 1 suspicious case and 3 inadequate specimens. Excellent correlation was noted between the cytologic and subsequent histopathologic diagnosis of 22 cases. Among 22 positive results, pathologic esamination showed 9 squamous cell carcinoma, 2 Adenosquamous cell carcinoma, 2 undifferentiated carcinoma, and 9 adenocarcinoma. There was no false positive case, and the sensitivity of the fine-needle aspiration was 84.6%. These data suggest that fine-needle aspiration cytology is a safe and effective method in diagnosis of gynecologic malignancies.