• 한국정보통신학회논문지
• /
• 제25권10호
• /
• pp.1279-1286
• /
• 2021

Electrocardiogram (ECG) 신호는 심장의 이상을 조기에 진단하기 위한 좋은 지표이다. ECG 신호는 사람마다 기준이 되는 정상 신호의 형태가 다르고, 진단에 많은 데이터가 필요하다. 본 논문에서는 ECG 신호 진단을 효율적으로 가속하기 위한 OpenCL을 기반 FPGA-GPU 혼합 계층 적응형 플랫폼을 제안한다. 플랫폼에서 MIT-BIH 부정맥 신호데이터의 19870개 ECG 신호를 진단한 결과 FPGA 가속기는 진단 시간이 1.15s로 소프트웨어로 실행했을 때보다 89.94% 감소하였고, 전력 소모는 84.0% 감소하였다. GPU 가속기는 실행 시간이 소프트웨어 대비 83.56% 감소한 1.87s였으며, 전력 소모는 62.3% 감소하였다. 제안하는 FPGA-GPU 혼합 플랫폼은 FPGA 가속기보다 진단 속도가 느리지만 GPU를 이용하여 상황에 따라 유연한 알고리즘을 동작할 수 있다.

• 한국환경복원기술학회지
• /
• 제23권6호
• /
• pp.187-200
• /
• 2020

Despite the eco-friendly route plan, road slopes were collapsed due to road construction, resulting in human and property damage. To solve this problem, the Ministry of Land, Transport and Maritime Affairs established guidelines in 2009 to conduct a test-bed for slope recording considering the conditions of the site and the ecological environment, and divided them into recording quality and economic evaluation. The following results were obtained by analyzing 183 construction methods of cut-soil sections at 60 sites from 2012 to 2018 for road slope afforestation research. Straw net+seedspray, vegetation media spray method 1T, 3T were used the most, and vegetation media spray method 3T(patent) was excellent in quality, and straw net+seedspray was excellent in economics. As a result of analyzing the market unit price and the construction unit price, vegetation media spray method submitted the construction unit price at up to 60% lower than the market unit price. As a result of the analysis of the key factors of the greening method evaluation, the economic assessment had the greatest influence on all evaluation items. Problems in the evaluation method of revegetation were first identified as problems in the allocation of points and secondly as problems in the evaluation criteria. As for the improvement of the economic assessment criteria, the method was proposed to evaluate the same method based on market unit price when the same method was constructed, and not to conduct an economic assessment if there was a difference in market unit price between methods, or to add weight to the scores. Based on the monitoring data of 60 road slopes, this study drew up problems and improvement measures. However, with regard to scoring, research on appropriate scoring is needed by examining the current status.

• 한국ITS학회 논문지
• /
• 제19권6호
• /
• pp.180-191
• /
• 2020

본 연구에서는 1999년부터 2014년까지 통계청에서 조사한 생활시간조사 자료를 활용하여 조사연도별로 평일 통행시간예산에 영향을 미치는 요인을 분석하여 이를 비교하고자 하였다. 먼저 인구 및 인구 및 사회경제지표, 비가정 활동시간 등을 비가정 활동시간 등을 고려한 통행시간에 관한 다중회귀모형을 구축하여 영향력을 분석하였다. 모형추정 결과, 가구특성, 개인특성, 비가정 활동시간 변수들이 통행시간에 유의하게 영향을 미치는 것으로 분석되었으며, 연도별로 영향력에 차이가 있음을 확인하였다. 또한, 통행시간과 비가정 활동시간간의 상관성을 고려하여 SUR모형을 구축하였으며, 독립변수들이 미치는 영향력을 시계열적으로 비교·분석하였다. 전반적으로 인구 및 사회경제지표가 통행시간은 물론 비가정 활동시간들에 지속적으로 영향을 미치고 있는 것으로 나타났다.

• 한국콘텐츠학회논문지
• /
• 제21권3호
• /
• pp.282-300
• /
• 2021

본 연구의 목적은 코로나19 확산으로 가족생활에서 급격한 변화가 일어난 1차 확산기에 블로그와 온라인 카페에 게시된 소셜 빅데이터를 분석하여 키워드를 파악하고, 게시글에 잠재된 주요 토픽을 발견하는 것이다. 강화된 사회적 거리두기가 처음 시행되었던 2020년 2월 23일부터 4월 19일까지 네이버와 다음의 블로그 및 카페에 게시된 글 중 '코로나'와 '가족' 또는 '코로나'와 '가정'이 함께 언급된 문서 총 351,734건을 분석하였다. 수집된 데이터는 전처리를 거쳐 텍스트 마이닝 기법으로 분석하였다. TF-IDF 가중치 값을 바탕으로 상위 100개 단어를 살펴보았으며, 잠재디리클레할당 방식의 토픽모델 분석을 통해 총 22개 토픽을 도출하고 토픽명을 부여하였다. 연구결과, 코로나19가 가족의 일상생활에 미친 전방위적 영향이 나타났으며, 특히 식생활, 주거생활, 여가생활, 종교생활, 자녀돌봄, 자녀교육, 가족관계, 가족의례 등에서 변화가 두드러졌다. 더불어, 가족 관련 국내 문헌에서는 잘 논의되지 않던 건강공동체로서의 가족을 시사하는 토픽도 등장하였다.

• 아태비즈니스연구
• /
• 제10권4호
• /
• pp.223-236
• /
• 2019

Governance can be defined as a concept that encompasses a series of processes including partner selection as well as control and coordination of collaboration to achieve common goals. The study examined efforts to mitigate the risks of opportunistic behaviors into partner selection, partner relationship, control mechanism, and conflict management. For cases that have achieved commercialization outputs with the participation of SMEs, data was collected and analyzed such as interviews with project managers and case records for seven months from October 2016. According to the analysis result, as the complexity increases, such as multilateral cooperation for the development of finished products, cooperation with a trusted partner rather than a partner who can perform a task well was preferred, and the process control was put ahead of the output control. Regarding the partner relationship, the relationship between the owner and the agent differed according to the point of view, and there was a lack of clear allocation of authority and responsibility as well as a reward for the result. In terms of the conflict management, most emphasis was on resolving conflicts or difficulties, and no attempt was made to utilize the positive aspects of the conflict. The structure of most industry-university-institute cooperation organizations is simply composed of the host and participating organizations, and the management regulations should be amended for companies, that put actual funds and use the outputs, to have the authority and responsibility as the owners, and be allowed to use the governance elements appropriately to take the lead as consumers.

• 한국정보통신학회논문지
• /
• 제23권11호
• /
• pp.1321-1329
• /
• 2019

소셜 정보망의 발달로 마케팅의 방법도 다양하게 변화되고 있다. 기존의 유명인, 경제적 지원 기반의 성공적인 마케팅방법론과 달리, 최근 인플루언서 기반 유튜브 마케팅이 큰 대세를 이루고 있다. 본 논문 에서는 처음으로 유튜브 양적 정보 및 댓글분석 기반 다각도 질적 분석을 활용하여 54개 이상의 유튜브 채널에서 인플루언서 특징을 추출하고 대표적인 주제들을 모델링하여 개인 맞춤형 영상 만족도 극대화는 물론 기업체가 새로운 아이템을 마케팅 할 때 기존의 인플루언서 특징을 참고하여 새로운 아이템의 영상을 제작하고 배포함으로써 성공적인 홍보 효과를 누릴 수 있도록 보조 수단 제공을 목적으로 한다. 유튜브 채널 별 다양한 영상의 모든 댓글을 각 문서로 가정하고 TF-IDF 및 LDA알고리즘을 적용하여 성능 극대화 향상을 보였다.

• 한국정보시스템학회지:정보시스템연구
• /
• 제30권4호
• /
• pp.277-301
• /
• 2021

Purpose The purpose of this study is to analyze the sentiment responses of the general public to non-face-to-face work using text mining methodology. As the number of non-face-to-face complaints is increasing over time, it is difficult to review and analyze in traditional methods such as surveys, and there is a limit to reflect real-time issues. Approach This study has proposed a method of the research model, first by collecting and cleansing the data related to non-face-to-face work among tweets posted on Twitter. Second, topics and keywords are extracted from tweets using LDA(Latent Dirichlet Allocation), a topic modeling technique, and changes for each section are analyzed through DTM(Dynamic Topic Modeling). Third, the complaints of non-face-to-face work are analyzed through the classification of positive and negative polarity in the COVID-19 section. Findings As a result of analyzing 1.54 million tweets related to non-face-to-face work, the number of IDs using non-face-to-face work-related words increased 7.2 times and the number of tweets increased 4.8 times after COVID-19. The top frequently used words related to non-face-to-face work appeared in the order of remote jobs, cybersecurity, technical jobs, productivity, and software. The words that have increased after the COVID-19 were concerned about lockdown and dismissal, and business transformation and also mentioned as to secure business continuity and virtual workplace. New Normal was newly mentioned as a new standard. Negative opinions found to be increased in the early stages of COVID-19 from 34% to 43%, and then stabilized again to 36% through non-face-to-face work sentiment analysis. The complaints were, policies such as strengthening cybersecurity, activating communication to improve work productivity, and diversifying work spaces.

• 한국프랜차이즈경영연구
• /
• 제13권4호
• /
• pp.1-12
• /
• 2022

Purpose: In 2020, the franchise industry accomplished a significant growth compared to the previous year, as the number of franchise companies increased by 9.0% while the number of franchise brands increased by 12.5%. Despite growth in size, the Korean franchise industry underwent many negative incidents, such as franchise ownership sales to private equity funds, that led to deterioration of businesses. From this point of view, this study aims to make various proposals to help policy makers develop franchise industry policies by analyzing trends of the current and previous presidential administrations' franchise policies and regulations using newspaper articles. Research design, data and methodology: A total of 7,439 articles registered in Naver API from February 25, 2013 to November 29, 2021 were extracted. Among them, 34 unrelated video articles were deleted, and a total of 7,405 articles from both administrations were used for analysis. The R package was used for word frequency analysis, word clouding, word correlation analysis, and LDA (Latent Dirichlet Allocation) topic modeling. Results: The keyword frequency analysis shows that the most frequently mentioned keywords during the previous administration include 'no-brand', 'major company', 'bill', 'business field', and 'SMEs', and those mentioned during the current administration include 'industry' and 'policy'. As a result of LDA topic modeling, 9 topics such as 'global startups' and 'job creation' from the previous administration, and 10 topics such as 'franchise business' and 'distribution industry' from the current administration were derived. The results of LDAvis showed that the previous administration operated a policy based on mutual growth of large and small businesses rather than hostile regulations in the franchise business, whereas the current administration extended the regulation related to franchise business to the employment sector. Conclusions: The analysis of past two administrations' franchise policy, it can be suggested that franchisors and franchisees may complement each other in developing the Fair Transactions in Franchise Business Act and achieving balanced growth. Moreover, political support is needed for sound development of franchisors. Limitations and future research suggestions are presented at the end of this study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권11호
• /
• pp.3738-3760
• /
• 2022

Due to increasing spectrum demand for new wireless devices applications, cooperative spectrum sensing (CSS) paradigm is the most promising solution to alleviate the spectrum shortage problem. However, in the interweave cognitive radio (CR) system, the inherent nature of CSS opens a hole to Byzantine attack, thereby resulting in a significant drop of the CSS security and efficiency. In view of this, a weighted differential sequential single symbol (WD3S) algorithm based on MATLAB platform is developed to accurately identify malicious users (MUs) and benefit useful sensing information from their malicious reports in this paper. In order to achieve this, a dynamic Byzantine attack model is proposed to describe malicious behaviors for MUs in an interweave CR system. On the basis of this, a method of data transmission consistency verification is formulated to evaluate the global decision's correctness and update the trust value (TrV) of secondary users (SUs), thereby accurately identifying MUs. Then, we innovatively reuse malicious sensing information from MUs by the weight allocation scheme. In addition, considering a high spectrum usage of primary network, a sequential and differential reporting way based on a single symbol is also proposed in the process of the sensing information submission. Finally, under various Byzantine attack types, we provide in-depth simulations to demonstrate the efficiency and security of the proposed WD3S.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• 제48권6호
• /
• pp.356-362
• /
• 2022

Objectives: A single-blinded randomized controlled trial was designed to compare and evaluate the effectiveness of the periotome and piezotome as aids for atraumatic extraction and its sequalae. Materials and Methods: The study sample comprised 48 teeth, equally allotted to the piezotome or periotome groups by random allocation, in participants aged 19-62 years. All samples in both groups had either complete tooth structure or intact roots without crowns and had mobility ≤grade II. Clinical parameters of operative duration, presence or absence of gingival laceration, reported operative and postoperative pain, and intake of analgesics following extraction were recorded. IBM SPSS software package version 22 was used for data entry and analysis. Results: The mean operation time was significantly (P≤0.05) longer in the piezotome group than in the periotome group. However, fewer gingival lacerations were observed with use of a piezotome than with a periotome, although no significant difference was observed. The piezotome group reported significantly (P≤0.05) higher visual analog scale (VAS) pain scores during the procedure and non-significantly higher scores thereafter until the third postoperative day. In the piezotome group, the dosage of analgesic was higher, although the periotome group had a higher percentage of participants who used analgesics postoperatively; however, these differences were not statistically significant. Conclusion: The present clinical trial favors the use of periotome over piezotome for atraumatic extraction due to shorter operating time, lower postoperative VAS pain scores, and lower dosage of analgesics despite the superior ability of the piezotome to prevent gingival laceration.