• 한국컴퓨터정보학회논문지
• /
• 제26권1호
• /
• pp.265-274
• /
• 2021

본 논문에서는 검색 키워드와 상품 상세정보를 활용한 Doc2vec 기반의 새로운 추천 모형을 제안한다. 지금까지 추천 시스템에 관한 많은 기존 연구에서는 고객의 구매 이력이나 평점 같은 정형 데이터만을 사용하는 협업 필터링(CF) 알고리즘에 기반한 추천 모델이 제안되었다. 그러나 CF에서 온라인 고객 리뷰와 같은 비정형 데이터를 사용하면, 보다 나은 추천결과를 도출할 수 있다. 이에 본 연구에서는 기존 연구에서 거의 활용되지 않았던 검색 키워드 정보와 상품 상세정보를 제품 추천에 활용할 것을 제안한다. 본 연구의 제안 모형은 고객이 구매한 상품에 대한 평점, 검색어, 상품 상세정보를 종합적으로 고려한 CF 알고리즘을 이용해 추천결과를 생성한다. 이 때 비정형 데이터로부터 정량적인 패턴을 추출하기 위한 방법으로는 Doc2vec이 적용된다. 실험 결과 제안 모형이 기존 추천 모형보다 더 나은 성능을 보이는 것을 알 수 있었고, 검색어 및 상품 상세정보가 추천에 유의한 영향을 미치는 것을 확인하였다. 본 연구는 고객의 온라인 행동 정보를 추천시스템에 적용하였다는 점과 전통적인 CF의 한계 중 하나인 콜드 스타트 문제를 완화하였다는 점에서 학술적 의의가 있다.

• 정보화연구
• /
• 제9권3호
• /
• pp.323-331
• /
• 2012

지식정보화 시대의 본격화와 함께 지식재산권, 그 중에서도 특허의 중요성이 더욱 커져가고 있다. 이에 따라 효율적인 특허정보 검색방법의 필요성이 높아지고 있지만, 기존의 특허검색 엔진은 불리언 모델을 기반으로 단어의 존재 여부만을 파악하는 방식으로 검색결과에 노이즈 데이터가 너무 많이 포함되어 특허 검색에 오랜 시간을 허비하게 만들므로 '전문검색가'들이 수동으로 찾아주고 있는 실정이다. 이에 본 논문에서는 기존의 일반적 문서검색과 특허검색과의 차이점을 밝히고, 기존 특허검색의 한계성을 분석한다. 나아가 특허검색에 특화된 효과적 방법론 제안하여 검색 키워드가 각 특허 문서 내에서 차지하는 중요도와 각 문서 내에서 키워드 사이의 관계성을 파악하고 이에 대한 랭킹을 정하여 키워드와 관계성이 높은 특허가 상위에 랭크하며 노이즈 데이터를 하위에 랭크 함으로써 검색 결과에서 노이즈 데이터의 비율을 대폭 줄이는 방법을 제안한다. 마지막으로 실험을 통하여 Kipris 검색 결과와 비교함으로써 제안한 방법론의 우수성을 입증하였다.

• 동의신경정신과학회지
• /
• 제34권1호
• /
• pp.13-21
• /
• 2023

Objectives: The aim of this study was to analyze correlations among searches of hwa-byung, depression, and suicide using big data. Methods: Keywords searches were performed using both Google Trends and Naver Data Lab on December 13, 2022. From 2016 to 2022, search results for keywords 'hwa-byung', 'depression', and 'suicide' were extracted with a score between 0 and 100 in terms of relative search popularity (RSP). Monthly time analysis, correlation analysis, and regional analysis were then conducted for these scores. Results: Regardless of the search period, RSP for both portal sites was in the order of 'suicide', 'depression', and 'hwa-byung'. Over time, search for 'depression' tended to increase in Google (slope: 0.0092), whereas search for 'hwa-byung' showed a slight increase in Naver (slope: 0.0024). Correlation coefficient for search terms 'depression' and 'suicide' was 0.3969 in Google Trends and 0.4459 in Naver Data Lab, showing clear positive correlations. On the other hand, there was little correlation between search results of 'hwa-byung' and 'depression' or between 'hwa-byung' and 'suicide'. However, compared to males, females showed higher positive associations between search results of 'hwa-byung' and 'depression' and between 'hwa-byung' and 'suicide'. Search terms 'depression' and 'suicide' showed high RSPs in most regions in South Korea. However, 'hwa-byung' had distinct regional differences in terms of RSP. Conclusions: Results of this study will help us understand Korean public's perception of the relevance of hwa-byung, depression, and suicide and plan future research in this topic. In addition, findings of this study may provide future public health implications for reducing the high suicide rate in Korea.

• 치위생과학회지
• /
• 제21권2호
• /
• pp.104-110
• /
• 2021

Background: Regardless of countries, the myth that rain makes the body ache has been worded in various forms, and a number of studies have been reported to investigate this. However, these studies, which depended on the patient's experience or memory, had obvious limitations. Google Trends is a big data analysis service based on search terms and viewing videos provided by Google LLC, and attempts to use it in various fields are continuing. In this study, we endeavored to introduce the 'value as a research tool' of the Google Trends, that has emerged along with technological advancements, through research on 'whether toothaches really occur frequently on rainy days'. Methods: Keywords were selected as objectively as possible by applying web crawling and text mining techniques, and the keyword "bi" meaning rain in Korean was added to verify the reliability of Google Trends data. The correlation was statistically analyzed using precipitation and temperature data provided by the Korea Meteorological Agency and daily search volume data provided by Google Trends. Results: Keywords "chi-gwa", "chi-tong", and "chung-chi" were selected, which in Korean mean 'dental clinic', 'toothache', and 'tooth decay' respectively. A significant correlation was found between the amount of precipitation and the search volume of tooth decay. No correlation was found between precipitation and other keywords or other combinations. It was natural that a very significant correlation was found between the amount of precipitation, temperature, and the search volume of "bi". Conclusion: Rain seems to actually be a cause of toothache, and if objective keyword selection is premised, Google Trends is considered to be very useful as a research tool in the future.

• 대한한의학회지
• /
• 제44권3호
• /
• pp.74-86
• /
• 2023

Objectives: The purpose of this study was to examine search trends and their meanings for major uninsuired Korean medicine (KM) treatments through analysis of an online portal search results. Methods: Keywords searches were performed using Naver Datalab on 4 July 2023. From January 2016 to June 2023, monthly relative search volume (RSV) for keywords 'pharmacopuncture', 'Chuna', and 'needle-embedding therapy', and 'herbal decoction' were extracted with a score between 0 and 100. For the obtained RSVs, longitudinal changes over time, characteristics according to sex and age group, and correlations between them were investigated. Results: The ranking of RSV for each keyword has changed from 'Chuna', 'herbal decoction', 'needle-embedding therapy', and 'pharmacopuncture' to 'Chuna', 'herbal decoction', 'pharmacopuncture', and 'needle-embedding therapy' after 2019. Overall, the RSV of needle-embedding therapy continuously decreased, while that of pharmacopuncture continuously increased. In 2019, a rapid increase in the RSV of Chuna was observed, and in 2020, a rapid increase in the RSV of herbal decoction was observed. There was a difference in the longitudinal change pattern of RSV for the keywords by age group. Importantly, in the elderly, changes in RSV were observed in a favorable pattern to KM treatment. Conclusion: Our findings enable estimation of the public's interest and its changes for the four uninsuired KM treatment, and can be used as basic data to strengthen health insurance coverage in Korea. Specifically, changes in interest in KM treatments according to sex and age can be referred to.

• 한국정보시스템학회지:정보시스템연구
• /
• 제30권1호
• /
• pp.123-149
• /
• 2021

Purpose The purpose of this study is to define the anti-ESG activities of companies recognized by media by reflecting ESG recently attracted attention. This study extracts keywords for ESG controversies through association rule mining. Design/methodology/approach A research framework is designed to extract keywords for ESG controversies as follows: 1) From DeepSearch DB, we collect 23,837 articles on anti-ESG activities exposed to 130 media from 2013 to 2018 of 294 listed companies with ESG ratings 2) We set keywords related to environment, social, and governance, and delete or merge them with other keywords based on the support, confidence, and lift derived from association rule mining. 3) We illustrate the importance of keywords and the relevance between keywords through density, degree centrality, and closeness centrality on network analysis. Findings We identify a total of 26 keywords for ESG controversies. 'Gapjil' records the highest frequency, followed by 'corruption', 'bribery', and 'collusion'. Out of the 26 keywords, 16 are related to governance, 8 to social, and 2 to environment. The keywords ranked high are mostly related to the responsibility of shareholders within corporate governance. ESG controversies associated with social issues are often related to unfair trade. As a result of confidence analysis, the keywords related to social and governance are clustered and the probability of mutual occurrence between keywords is high within each group. In particular, in the case of "owner's arrest", it is caused by "bribery" and "misappropriation" with an 80% confidence level. The result of network analysis shows that 'corruption' is located in the center, which is the most likely to occur alone, and is highly related to 'breach of duty', 'embezzlement', and 'bribery'.

• International Journal of Internet, Broadcasting and Communication
• /
• 제8권2호
• /
• pp.1-22
• /
• 2016

While many organizations believe that cloud computing has the potential to reduce operational cost by abstracting capital assets like data storage center and processing systems into a readily on demand available and affordable operating expenses, still many of these organizations are not aware of the factors determining the performance of cloud computing technology. This paper provides a systematic literature review focusing on the factors determining the performance of cloud computing. In trying to come up with this review, the following sources were searched for relevant articles: ScienceDirect, Scientific.Net, ACMDigital Library, IEEE Xplore, Springer, World Scientific Journal, Wiley Online Library, Academic Search Premier (via EBSCOHost) and EdITLib (Education & Information Technology Digital Library). In first search strategy, approximately 100 keywords related to the research domain like; "Cloud Computing" and "Cloud Services" were used. In second search strategy, 65 keywords more related to the research domain were selected. In the third search strategy, the primary materials were identified and classified according to the paper types (Journal or Conference), year of publication and so on. Based on this study, twenty (20) factors were found that determine the performance of cloud computing. The IT organization needs to consider these twenty (20) factors in order to adopt cloud computing.

• 정보처리학회논문지D
• /
• 제15D권4호
• /
• pp.485-494
• /
• 2008

인터넷이 광범위하게 활용됨에 따라 검색 키워드, 멀티미디어 객체, 웹 페이지, 블로그 등의 다양한 웹 객체들이 크게 증가하고 있다. 이러한 웹 객체들의 인기도는 시간에 따라 변화하며, 그러한 웹 객체 인기도의 시간적 패턴에 대한 마이닝이 여러 가지 웹 응용에 필요한 중요한 연구 과제가 되고 있다. 예를 들어, 검색 키워드에 대한 인기도 패턴의 분석은 앞으로 인기가 높아질 키워드를 미리 예측할 수 있게 하여 광고주들에게 키워드를 판매하기 위한 가격을 결정하는 데에 중요한 자료가 될 수 있다. 하지만, 웹 객체 인기도가 시간에 따라 변화하고 웹 객체의 개수가 매우 방대하다는 특성으로 인하여 웹 객체 인기도에 대한 분석은 매우 어려운 문제이다. 본 논문에서는 웹 객체 인기도의 시간적 패턴을 마이닝하기 위한 효율적인 알고리즘을 제안한다. 본 논문은 웹 객체 인기도를 시계열로 표현하고, 두 웹 객체 인기도 간의 유사성을 측정하기 위하여 gap 척도를 제안한다. gap 척도의 효율적인 계산을 위하여 FFT를 활용한 알고리즘을 제안하고, 밀도기반 클러스터링 알고리즘을 이용하여 유사한 인기도 추세를 갖는 웹 객체들의 클러스터를 생성한다. 본 논문에서는 웹 객체 인기도가 특정 분포를 따르거나 주기적이라고 가정하지 않는다. Google Trends 웹 사이트로부터 구한 검색 키워드 인기도를 이용한 실험을 통하여, 제안된 알고리즘이 실세계 응용에서 유용함을 보인다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권2호
• /
• pp.179-184
• /
• 2024

Franchise are now a red ocean in Food industry and they need to find other options to appeal for their product, the uprising content, food tech. The franchises are working on R&D to help franchisees with the operations. Through this paper, we analyze the franchise interest on food tech and to help find the necessity of development for franchisees who are in needs with hand, not of human, but of technology. Using Textom, a big data analysis tool, "franchise" and "food tech" were selected as keywords, and search frequency information of Naver and Daum was collected for a year from 01 January, 2023 to 31 December, 2023, and data preprocessing was conducted based on this. For the suitability of the study and more accurate data, data not related to "food tech" was removed through the refining process, and similar keywords were grouped into the same keyword to perform analysis. As a result of the word refining process, a total of 10,049 words were derived, and among them, the top 50 keywords with the highest relevance and search frequency were selected and applied to this study. The top 50 keywords derived through word purification were subjected to TF-IDF analysis, visualization analysis using Ucinet6 and NetDraw programs, network analysis between keywords, and cluster analysis between each keyword through Concor analysis. By using big data analysis, it was found out that franchise do have interest on food tech. "technology", "franchise", "robots" showed many interests and keyword "R&D" showed that franchise are keen on developing food tech to seize competitiveness in Franchise Industry.

• 한국디지털건축인테리어학회논문집
• /
• 제12권2호
• /
• pp.53-60
• /
• 2012

In this paper, we present new methods for identifying keywords for foresight topics that utilize the internet and textmining techniques to draw objective and quantified information that support experts' qualitative opinions and evaluations in foresight. Furthermore, by applying this fabricated procedure, we have derived keywords to analyze priorities in architectural engineering. Not much difference between qualitative methods of experts and quantitative methods such as text mining has been observed from comparison between technologies derived via qualitative method from "The Science Technology Vision" (control group). Therefore, as a quantitative tool useful for drawing keywords for foresight, textmining can supplement quantitative analysis by experts. In addition, depending on the level and type of raw data, text mining can bring better results in deriving foresight keywords. For this reason, research activities accommodating Internet search results and the development of textmining methods for analyzing current trends are in demand.