• Korean Association of Arts Management
• /
• no.50
• /
• pp.5-31
• /
• 2019

The main purpose of this study was to use Deep Learning based Topic Modeling and Semantic Network Analysis to examine research trend of arts management-related papers in korea. For this purpose, research subjects such as 'The Journal of Cultural Policy', 'The Journal of Cultural Economics', 'The Journal of Culture Industry', 'The Journal of Arts Management', and 'The Journal of Human Content', which are the registered journal of the National Research Foundation of Korea directly or indirectly related to arts management field. From 1988 to 2017, a total of 2,110 domestic journals' signature, abstract, and keyword were analyzed. We tried Big Data analysis such as Topic Modeling and Semantic Network Analysis to examine changes in trends in arts management. The analysis program used open software R and standard statistical software SPSS. Based on the results of the analysis, the implications and limitations of the study and suggestions for future research were discussed. And the potential for development of convergent research such as Arts & Artificial Intelligence and Arts & Big Data.

• Proceedings of the Korea Contents Association Conference
• /
• 2019.05a
• /
• pp.105-106
• /
• 2019

본 연구는 토픽모델링 기법을 국가의 연구개발 프로젝트에 적용하여 클러스터링하고 네트워크 분석을 통해 개별 클러스터와 R&D프로젝트를 평가하는 것에 관한 것이다.

• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.282-300
• /
• 2021

The purpose of this study was to explore what social media posts tell us about family life during the COVID-19 pandemic by examining the keywords and topics underlying posts on blogs and online forums. Our criteria for web crawling were (a) blog and forum posts on Naver and Daum, the top portal sites in Korea, (b) posts between February 23 and April 19, 2020, the period of the first heightened social distancing orders, and (c) inclusion of "COVID" and "family" or "COVID" and "home." We analyzed 351,734 posts using TF-IDF values and topic modeling based on latent Dirichlet allocation. We identified and named 22 topics including COVID-19 prevention, family infection, family health, dietary life and changes, religious life, stuck at home, postponed school year, family events, travel and vacations, concerns about family and friends, anxiety and stress, disaster and damage, COVID-19 warning text messages, family support policies, Shin-cheon-ji and Daegu. The results show that COVID-19 impacted various domains of family life including health, food, housing, religion, child care, education, rituals, and leisure as well as relationships and emotions.

• Journal of Internet Computing and Services
• /
• v.25 no.3
• /
• pp.27-33
• /
• 2024

The environment of academic research is continuously changing due to the increase of information, which raises the need for an effective way to analyze and organize large amounts of documents. In this paper, we propose Performance Improvement of Topic Modeling using BART(Bidirectional and Auto-Regressive Transformers) based Document Summarization. The proposed method uses BART-based document summary model to extract the core content and improve topic modeling performance using LDA(Latent Dirichlet Allocation) algorithm. We suggest an approach to improve the performance and efficiency of LDA topic modeling through document summarization and validate it through experiments. The experimental results show that the BART-based model for summarizing article data captures the important information of the original articles with F1-Scores of 0.5819, 0.4384, and 0.5038 in Rouge-1, Rouge-2, and Rouge-L performance evaluations, respectively. In addition, topic modeling using summarized documents performs about 8.08% better than topic modeling using full text in the performance comparison using the Perplexity metric. This contributes to the reduction of data throughput and improvement of efficiency in the topic modeling process.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.551-552
• /
• 2017

하나의 프로젝트에는 다양한 기능과 역할을 가진 소스코드가 존재한다. 그러나 기존 정적 분석 도구들은 이러한 특성을 고려하지 않고, 모든 소스코드에 동일한 탐색 정책과 우선순위를 적용하고 있다. 본 연구에서는 오픈소스 프로젝트로부터 수집한 소스코드들을 토픽모델링을 이용하여 특정 토픽으로 분류하고, 분류된 토픽에 해당되는 코드 안에서 높은 영향력을 갖는 잠재결함(Potential Bug)의 특징을 분석하였다. 이 결과를 바탕으로 개발자에게 개발 중인 소스코드의 특성에 따라 어떤 잠재결함에 더 우선순위를 두어야 하는지에 대한 지침을 제공할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.464-467
• /
• 2022

기업의 사회적 책임에 대한 요구가 높아짐에 따라 기업의 지속 가능 경영 보고서 발간은 증가 추세를 보이고 있다. 그러나 이전까지의 연구는 지속가능성 및 기업의 재무적, 비재무적 연관성에 초점이 맞춰져 있었으며, 전통적인 토픽 모델링 기법만을 제한적으로 사용한다는 한계를 보였다. 본 연구에서는 Transformer 기반의 맥락을 고려한 토픽 모델링 기법을 도입하여 다양한 이해관계자 측면에서 이용 가능한 25 개의 주제를 도출하였다. 또한 동적 토픽 모델링(Dynamic Topic Modeling)을 통해 주제의 변화를 시계열적으로 파악했다.

• Journal of Internet Computing and Services
• /
• v.22 no.5
• /
• pp.47-55
• /
• 2021

Analysis of research trends in specific subject areas is performed by examining related topics and subject changes by using topic modeling techniques through keyword extraction for most of the literature information (paper, patents, etc.). Unlike existing research methods, this paper extracts topics related to the research topic using the LDA topic modeling technique for the project information of national R&D projects provided by the National Science and Technology Knowledge Information Service (NTIS) in the field of artificial intelligence. By analyzing these topics, this study aims to analyze research topics and investment directions for national R&D projects. NTIS provides a vast amount of national R&D information, from information on tasks carried out through national R&D projects to research results (thesis, patents, etc.) generated through research. In this paper, the search results were confirmed by performing artificial intelligence keywords and related classification searches in NTIS integrated search, and basic data was constructed by downloading the latest three-year project information. Using the LDA topic modeling library provided by Python, related topics and keywords were extracted and analyzed for basic data (research goals, research content, expected effects, keywords, etc.) to derive insights on the direction of research investment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.3
• /
• pp.661-670
• /
• 2021

Research trends can be usefully used to determine the importance of research topics by period, identify insufficient research fields, and discover new fields. In this study, research trends of urban spaces, where various problems are occurring due to population concentration and urbanization, were analyzed by topic modeling. The analysis target was the abstracts of papers listed in the Korea Citation Index (KCI) published between 2002 and 2019. Topic modeling is an algorithm-based text mining technique that can discover a certain pattern in the entire content, and it is easy to cluster. In this study, the frequency of keywords, trends by year, topic derivation, cluster by topic, and trend by topic type were analyzed. Research in urban regeneration is increasing continuously, and it was analyzed as a field where detailed topics could be expanded in the future. Furthermore, urban regeneration is now becoming a regular research field. On the other hand, topics related to development/growth and energy/environment have entered a stagnation period. This study is meaningful because the correlation and trends between keywords were analyzed using topic modeling targeting all domestic urban studies.

• Smart Media Journal
• /
• v.13 no.1
• /
• pp.36-44
• /
• 2024

Metaverse is attracting attention as the development of virtual environment technology and the emergence of untact culture due to the COVID-19 pandemic. In this study, by analyzing users' reviews on the "Zepeto" application, which has recently attracted attention as a metaverse service, we tried to confirm changes in the requirements for the metaverse after the COVID-19 pandemic. To this end, 109,662 reviews of "Zepeto" applications written on the Google Play Store from September 2018 to March 2023 were collected, topics were extracted using LDA topic modeling technique, and topics were analyzed using the Causal Impact technique to examine how topics changed before and after based on "March 11, 2020" when the COVID-19 pandemic was declared. As a result of the analysis, five topics were extracted: application functional problems (topic1), security problems (topic 2), complaints about cryptocurrency (Zem) in the application (topic 3), application performance (topic 4), and personal information-related problems (topic 5). Among them, it was confirmed that security problems (topic 2) were most affected by the COVID-19 pandemic.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.884-887
• /
• 2010

최근 검색엔진의 효율성을 향상시키고 검색결과에 있어서 사용자들의 요구사항을 충족시키기 위한 연구들이 활발히 수행되고 있으며, 많은 방법론들이 제시되고 있다. 이는 방대한 정보 속에서 사용자의 검색 의도에 맞는 정보를 효과적으로 제공하는 것을 그 목표로 한다. 특히 본 논문에서는 검색하고자 하는 토픽별 사용자의 인적 속성들이 미치는 영향력을 기반으로 사용자간 소셜 관계 지수(SRV : Social Relationship Value)를 산정하는 방법을 제안한다. 소셜 관계 지수란 인간의 내재적인 특성을 수치로 산정한 것으로, 웹 사용자들에게 있어서는 검색 성향의 유사정도와 직결된다. 따라서 검색하고자 하는 토픽별 개인 성향의 유사정도를 수치로 부여하고 유사성이 높은 사람들의 검색 정보를 이용하면 사용자에 보다 만족된 검색결과를 제공할 수 있다. 본 연구에서는 구글 디렉터리(Google directory)의 정제된 각 토픽별 하위 범주(category)에 대해 선택 결과가 같은 사람들을 대상으로 인적 속성을 분석하고, 그 영향력을 가중치로 적용해 산정된 소셜 관계 지수와 사용자들의 검색 패턴을 비교 하였다. 그 결과 특정인을 기준으로 소셜 관계 지수가 높은 사람들의 검색 패턴이 매우 유사함을 확인 하였다. 이를 통해 토픽별 개인 간 연결 강도가 강할수록, 즉 유사성이 높은 사용자간에는 검색 패턴 또한 유사함을 검증 할 수 있었다.