댓글 분석을 통한 19대 한국 대선 후보 이슈 파악 및 득표율 예측

서대호(Dae-Ho Seo),김지호(Ji-Ho Kim),김창기(Chang-Ki Kim) 한국지능정보시스템학회 2018 지능정보연구 Vol.24 No.3

인터넷의 일상화와 각종 스마트 기기의 보급으로 이용자들로 하여금 실시간 의사소통이 가능하게 하여 기존의 커뮤니케이션 양식이 새롭게 변화되었다. 인터넷을 통한 정보주체의 변화로 인해 데이터는 더욱 방대해져서 빅데이터라 불리는 정보의 초대형화를 야기하였다. 이러한 빅데이터는 사회적 실제를 이해하기 위한 새로운 기회로 여겨지고 있다. 특히 텍스트 마이닝은 비정형 텍스트 데이터를 이용해 패턴을 탐구하여 의미있는 정보를 찾아낸다. 텍스트 데이터는 신문, 도서, 웹, SNS 등 다양한 곳에 존재하기 때문에 데이터의 양이 매우 다양하고 방대하여 사회적 실제를 이해하기 위한 데이터로 적합하다. 본 연구는 한국 최대 인터넷 포털사이트 뉴스의 댓글을 수집하여 2017년 19대 한국 대선을 대상으로 연구를 수행하였다. 대선 선거일 직전 여론조사 공표 금지기간이 포함된 2017년 4월 29일부터 2017년 5월 7일까지 226,447건의 댓글을 수집하여 빈도분석, 연관감성어 분석, 토픽 감성 분석, 후보자 득표율 예측을 수행하였다. 이를 통해 각 후보자들에 대한 이슈를 분석 및 해석하고 득표율을 예측하였다. 분석 결과 뉴스 댓글이 대선 후보들에 대한 이슈를 추적하고 득표율을 예측하기에 효과적인 도구임을 보여주었다. 대선 후보자들은 사회적 여론을 객관적으로 판단하여 선거유세 전략에 반영할 수 있고 유권자들은 각 후보자들에 대한 이슈를 파악하여 투표시 참조할 수 있다. 또한 후보자들이 빅데이터 분석을 참조하여 선거캠페인을 벌인다면 국민들은 자신들이 원하는 바가 후보자들에게 피력, 반영된다는 것을 인지하고 웹상에서 더욱 적극적인 활동을 할 것이다. 이는 국민의 정치 참여 행위로써 사회적 의의가 있다. With the everyday use of the Internet and the spread of various smart devices, users have been able to communicate in real time and the existing communication style has changed. Due to the change of the information subject by the Internet, data became more massive and caused the very large information called big data. These Big Data are seen as a new opportunity to understand social issues. In particular, text mining explores patterns using unstructured text data to find meaningful information. Since text data exists in various places such as newspaper, book, and web, the amount of data is very diverse and large, so it is suitable for understanding social reality. In recent years, there has been an increasing number of attempts to analyze texts from web such as SNS and blogs where the public can communicate freely. It is recognized as a useful method to grasp public opinion immediately so it can be used for political, social and cultural issue research. Text mining has received much attention in order to investigate the public`s reputation for candidates, and to predict the voting rate instead of the polling. This is because many people question the credibility of the survey. Also, People tend to refuse or reveal their real intention when they are asked to respond to the poll. This study collected comments from the largest Internet portal site in Korea and conducted research on the 19th Korean presidential election in 2017. We collected 226,447 comments from April 29, 2017 to May 7, 2017, which includes the prohibition period of public opinion polls just prior to the presidential election day. We analyzed frequencies, associative emotional words, topic emotions, and candidate voting rates. By frequency analysis, we identified the words that are the most important issues per day. Particularly, according to the result of the presidential debate, it was seen that the candidate who became an issue was located at the top of the frequency analysis. By the analysis of associative emotional words, we were able to identify issues most relevant to each candidate. The topic emotion analysis was used to identify each candidate`s topic and to express the emotions of the public on the topics. Finally, we estimated the voting rate by combining the volume of comments and sentiment score. By doing above, we explored the issues for each candidate and predicted the voting rate. The analysis showed that news comments is an effective tool for tracking the issue of presidential candidates and for predicting the voting rate. Particularly, this study showed issues per day and quantitative index for sentiment. Also it predicted voting rate for each candidate and precisely matched the ranking of the top five candidates. Each candidate will be able to objectively grasp public opinion and reflect it to the election strategy. Candidates can use positive issues more actively on election strategies, and try to correct negative issues. Particularly, candidates should be aware that they can get severe damage to their reputation if they face a moral problem. Voters can objectively look at issues and public opinion about each candidate and make more informed decisions when voting. If they refer to the results of this study before voting, they will be able to see the opinions of the public from the Big Data, and vote for a candidate with a more objective perspective. If the candidates have a campaign with reference to Big Data Analysis, the public will be more active on the web, recognizing that their wants are being reflected. The way of expressing their political views can be done in various web places. This can contribute to the act of political participation by the people.

HTM 기반의 주식가격 연속 예측 시스템 개발

서대호(Dae-Ho Seo),배선갑(Sun-Gap Bae),김성진(Sung-Jin Kim),강현석(Hyun-Syug Kang),배종민(Jong-Min Bae) 한국멀티미디어학회 2011 멀티미디어학회논문지 Vol.14 No.9

텍스트 마이닝 기법 기반 국내 홈 퍼니싱 브랜드의 소비자 인식 및 동향 비교ㆍ분석 : IKEA와 한샘을 중심으로

서대호(주저자) ( Seo Dae-ho(주저자) ),한지은(교신저자) ( Han Jieun(교신저자) ) 디자인융복합학회 2018 디자인융복합연구 Vol.17 No.6

최근 국내 가구시장은 IKEA가 국내로 진출함에 따라 홈 퍼니싱에 대한 높은 관심을 불러일으켰으며 지속적인 매출 성장을 이어가고 있다. 하지만 현재 이렇게 성장 가능성이 주목되는 국내 홈 퍼니싱 시장에 대한 사용자의 인식 및 동향에 대한 실증적인 분석이 많지 않은 실정이다. 이에 본 연구는 국내 가구시장에 있어 IKEA와 한샘에 대한 소비자의 인식 및 동향을 비교 분석해 보는 것이 시기적절하다고 판단하였다. IKEA와 한샘에 대한 사용자의 인터넷 게시물을 기반으로 텍스트 마이닝 기법을 적용하여 키워드를 비교함으로써 두 브랜드에 대한 소비자의 인식 및 동향을 비교 분석해 보고자 한다. 연구결과 IKEA에 대한 7가지, 한샘에 대한 5가지 토픽과 핵심 키워드를 도출했으며 브랜드 이미지, 가구디자인, 사용성, 서비스의 관점에서 소비자의 인식 및 마케팅 전략방향까지 제안하였다. 이와 같은 분석을 통해 현재 국내 홈 퍼니싱 시장의 동향과 사용자 관점에서의 인식을 통찰하면서 기존의 문헌 분석 방법의 한계를 보완하는 텍스트 마이닝 기법의 활용을 제안하고자 한다. As IKEA have launched the domestic market in South Korea, the furniture market has attracted high interest in home furnishing and continues to grow its sales continuously. However, there is not much empirical analysis on user perceptions of the domestic home furnishing brand. Therefore, this study decided that it is timely to compare consumers' perceptions and trends of IKEA and HANSSEM in domestic furniture market. We implemented a topic modeling analysis with the text mining technique based on the user 's postings on Twitter. As a result, we have identified 7 topics about IKEA, 5 topics about Hanssem and keywords, and suggested the consumer's perception and marketing strategy from the viewpoint of brand image, design, usability and service. Through this study, we not only be able to gain insight into the current trends and user perspectives of the domestic home furnishing market, but also suggest the use of text mining techniques that complement the limitations of current literature review methods in design field.

부품 이미지 인식을 위한 HTM 네트워크 훈련 시스템 개발

이대한(Dae-Han Lee),배선갑(Sun-Gap Bae),서대호(Dae-Ho Seo),강현석(Hyun-Syug Kang),배종민(Jong-Min Bae) 한국멀티미디어학회 2010 멀티미디어학회논문지 Vol.13 No.11

텍스트 마이닝을 활용한 국외 현대무용 연구 트렌드 변화 분석

김하연 ( Kim Ha-yeon ),서대호 ( Seo Dae-ho ) 한국예술교육학회 2020 예술교육연구 Vol.18 No.1

식지 않은 K-POP 한류열풍으로 인해 K-POP댄스의 인기는 높아지고 있다. 이에 따라 다양한 댄스장르에 대한 관심도가 급증하게 되었고 대중예술과 순수예술에 대한 연구가 많이 시도되고 있다. 그러나 순수예술의 세부전공에 대한 학술적 연구는 아직 까지 미흡한 실정이다. 세부전공 중 다른 장르에 비해 자유로운 표현방식으로 주목받고 있는 현대무용의 역사는 국외에서부터 시작되었고 현재 까지 국외 현대무용 공연시장의 영향을 많이 받고 있다. 따라서 국내 현대무용 분야의 체계화된 학문적 이론 구축과 공연시장 활성화를 위해서는 현시점에서 국외 현대무용계 연구 트렌드 변화를 살펴보고 미래를 예측해보는 것은 굉장히 의미 있는 연구가 될 것이다. 따라서 본 연구는 WOS에 30년간 축적된 현대무용 국외 학술자료를 1기~3기로 나누어 연구 트렌드를 탐색하고자 하였다. 연구방법은 빈도분석, TF-IDF분석, 네트워크분석, 토픽모델링분석을 활용하였다. 본 연구를 통해 향후 국내 현대무용학자들과 현장 전문가들이 현대무용발전을 위해 보다나은 환경을 만들어 가는데 도움이 될 것으로 예측한다. The popularity of K-POP dance is increasing. As a result, interest in various dance genres has increased rapidly, and research on popular art and pure art is being attempted a lot. However, academic studies on the pure arts are still insufficient. The history of contemporary dance began from abroad and has been influenced by the international dance performance market. Therefore, in order to construct a systematic academic theory and revitalize the performance market in the domestic contemporary dance field, it will be very meaningful to look at the changes in the research trends of the contemporary dance industry to predict the future. Therefore, this study attempted to explore the research trends by dividing the foreign articles accumulated in the WOS into 3 periods. The research method was frequency analysis, TF-IDF analysis, network analysis, and topic modeling analysis. This study is expected to help domestic contemporary dancers and field experts to create a better environment for modern dance development.

제초제 paraquat와 bentazon의 세포독성과 3-methylcholanthrene의 독성경감효과

임요섭(Yo Sup Rim),서대호(Dae Ho Seo),한두석(Du Seok Han) 한국독성학회 2001 Toxicological Research Vol.17 No.2

This study were carried out to investigate cytotoxicity oj paraquat and bentazon that is scattering to farm products were essential for human diet and compensatory effects of 3-methylcholanthrene (3-MC) in vitro and in vivo. In vitro, The 5.0×10⁴cell/ml of NIH 3T3 fibroblast in each well of 24 multidish were cultured. After 24 hours, the cells were treated with solution of paraquat and bentazon (1, 25, 50, 100 J.1M respectively). After the NIH 3T3 Jib rob last of all groups were cultured in same condition for 48 hours, Sulfohordamin B Protein (SRB) assay were performed to evaluate the cytotoxicity of cell organelles. Paraquat and bentazon SRB50 were 1860.73 J.μM, 1913.38 J.μM respectively. In vivo, Sprague Dawley male rats divided into paraquat and bentazon only administered group and simultaneous application group of paraquat and bentazon and 3-MC. At 30 min. and I, 3, 6, 12,24,48 and 96 hrs. interval after each treatment, the animals were sacrificed by decapitation and kidney were immediately removed, immersed in fixatives, and processed with routine method for light microscopic study. Paraffin sections were stained with H-E, PAM, and PAS. Under the light microscope, atrophic change of renal corpuscles were frequently observed from 3 hrs after paraquat and bentazon treatment. The increase of the mesangium was apparent 12 hrs later after paraquat and bentazon treatment. Necrotic changes of the epithelium and loss of brush border of proximal tubules were most severe at 48 hrs after paraquat and bentazon treatment, respectively. In contrast there were no evidences of the toxic effects on renal tissues at 48hrs in paraquat and bentazon plus 3-MC treated groups.

HTM 기반의 소리 인식을 이용한 부품의 양ㆍ불량 판별 시스템

배선갑(Sun-Gap Bae),한창영(Chang-Young Han),서대호(Dae-Ho Seo),김성진(Sung-Jin Kim),배종민(Jong-Min Bae),강현석(Hyun-Syug Kang) 한국멀티미디어학회 2010 멀티미디어학회논문지 Vol.13 No.10

어성초 헥산 추출물이 암세포주에 미치는 항암효과

서대호,한두석 원광대학교 치의학연구소 2003 圓光齒醫學 Vol.12 No.1

In order to develop cancer agents which indicate weak side effects and strong cancer activity, cytotoxicity of Houttuynia cordata extracts against human oral epithelioid carcinoma cells (KB) and human skin melanoma cells (SK-MEL-3) was evaluated by MTT assay. KB and SK-MEL-3 cells were cultured in RPMI 1640 media containg 10% fetal bovine serum, antibiotics and fungizone. After incubation for 24 hours, the cells were treated with Houttuyunia cordata extracts by dose dependent manner for 48 hours under the same condition. Also, SRB assay for quantity of nucleaic protein, PAS reaction for glycoprotein in cytoplasm and XTT assay for extracellular laminin have done by Elisa reader (spectra Max 250, USA) and PAS stain, respectively. The microscopic study was carried out to observe morphological changes and PAS positive reactions in KB and SK-MEL-3. The results were as follows ; 1. The MTT absorbances against KB and SK-MEL-3 in 10^-2 ㎎/㎖ of hexane extracts were significantly decreased to 77.6% and 78.4%, respectively. 2. The SRB absorbance against KB and SK-MEL-3 in 10^-2 ㎎/㎖ of hexane extracts were significantly decreased to 79.2% and 79.5%, respectively. 3. In 10^-2 ㎎/㎖ of hexane extract, PAS reaction of KB cells was changed to negative and PAS positive cells of SK-MEL-3 were decrease to below 30%. 4. The XTT absorbances against KB and SK-MEL-3 in 10^-2 ㎎/㎖ of hexane extract were significantly decreased to 76.1% and 78.0%, respectively. 5. The morphological change were observed and number of cells were decreased against the cells in 10^-2 ㎎/㎖ of hexane extract. These results suggest that hexane extract of Houttuynia cordata has shown inhibitory effects in the proliferation, differentiation and metastasis against KB and SK-MEL-3.

정보시스템 부서 혁신을 위한 변화관리에 관한 연구 : C 의료원 사례를 중심으로

장윤희,이재범,서대호 서강대학교 경영학연구원 2000 서강경영논총 Vol.11 No.2