http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
I/O 부하와 데이터 지역성을 고려한 데드라인 기반 맵리듀스 스케쥴링 기법
황재민(Jaemin Hwang),김천중(Cheonjung Kim),오현교(Hyunkyo Oh),임종태(Jongtae Lim),복경수(Kyoungsoo Bok),유재수(Jaesoo Yoo) 한국콘텐츠학회 2014 한국콘텐츠학회논문지 Vol.14 No.12
본 논문에서는 데드라인 내에 잡을 완료시키기 위한 맵리듀스 스케쥴링 기법을 제안한다. 제안하는 기법은 제출된 잡들을 제한시간 내에 처리하기 위해 데이터 지역성 만족 여부를 확인하고 I/O 부하 및 데드라인 만족 여부를 고려한다. I/O 부하가 존재하는 노드에서 잡을 수행할 경우 복제본 노드의 데이터를 활용하여 잡 태스크 처리 속도를 향상시킨다. 또한, 잡 예상 완료 시간이 데드라인을 초과했음에도 가용 노드가 발생하지 않을 경우 데드라인에 여유가 있는 잡의 태스크를 지연시켜 잡의 완료 시간을 단축시킨다. 제안하는 기법의 우수성을 입증하기 위해 기존 연구와 성능 평가를 수행한다. In this paper, we propose a mapreduce scheduling scheme to complete jobs within a deadline. The proposed scheme first checks data locality and considers I/O load and deadline to complete the submitted jobs within a constraint time. When a job is processed in a node with I/O load, the data of replica nodes are utilized to enhance the job processing time. In addition, if avaliable nodes do not exist in spite that the expected completion time of the job is over the deadline, the job with the most deadline is delayed and then the urgent job is executed to reduce the job completion time. To show the efficiency of the proposed method, it is compared with the existing method through performance evaluation.
도로 네트워크에서 기준 궤적을 기반으로 간선간의 유사성을 고려하는 근사 궤적 클러스터링
이석희(Seokhee Lee),김천중(Cheonjung Kim),곽윤식(Yoonsik Kwak),강형일(Hyungil Kang),고대식(Daesik Ko),송석일(Seokil Song) 한국정보기술학회 2013 한국정보기술학회논문지 Vol.11 No.3
Recently, proposed trajectory clustering methods for road network has been focused on finding out frequent single paths. However, with considering the similarity between adjacent paths finding the set of frequent similar paths are also important. In this paper, we propose a clustering method for finding out frequent similar paths by enhancing the existing trajectory clustering methods in road network. Consequently, we present an approximate similarity measuring method between two paths. Then, based on the similarity measuring method, we propose a trajectory clustering method that finds out frequent similar paths. Finally, we implement our proposed clustering method based on RDBMS with SQL interface, and present the experimental results to show our proposed method works well.
대규모 RDF 데이터의 특성을 고려한 효율적인 색인 기법
김기연(Kiyeon kim),윤종현(Jonghyeon Yoon),김천중(Cheonjung Kim),임종태(Jongtae Lim),복경수(Kyoungsoo Bok),유재수(Jaesoo Yoo) 한국콘텐츠학회 2015 한국콘텐츠학회논문지 Vol.15 No.1
본 논문에서는 RDF 데이터 특성을 고려하여 대규모 데이터에 대한 질의 처리를 향상시키기 위한 새로운 색인 기법을 제안한다. 제안하는 기법은 RDF 트리플 중 주어와 술어의 값이 중복적으로 사용되는 특징을 이용하여 주어와 목적어를 S-O 색인으로 구성한다. 또한, 트리플 중 상대적으로 가장 적은 수의 값을 갖고 있는 술어는 별도의 P 색인으로 구성하여 총 색인의 크기를 최소화한다. 술어를 포함한 질의 요청시 크기가 작은 P 색인을 우선 검색하고 술어를 포함하지 않은 질의 요청에 대해서는 S-O 색인을 우선 검색한다. 성능평가를 통해 제안하는 기법이 기존 기법에 비해 질의처리 속도 관점에서 성능이 우수함을 보인다. In this paper, we propose a new RDF index scheme considering the characteristics of large scale RDF data to improve the query processing performance. The proposed index scheme creates a S-O index for subjects and objects since the subjects and objects of RDF triples are used redundantly. In order to reduce the total size of the index, it constructs a P index for the relatively small number of predicates in RDF triples separately. If a query contains the predicate, we first searches the P index since its size is relatively smaller compared to the S-O index. Otherwise, we first searches the S-O index. It is shown through performance evaluation that the proposed scheme outperforms the existing scheme in terms of the query processing time.
Kang Hyeon Kim,Kyung Wook Wee,CheonJung Kim,Don Hur,Jeong Hoon Lee,Yong Kyoung Yoo 대한의용생체공학회 2022 Biomedical Engineering Letters (BMEL) Vol.12 No.2
Field effect transistor (FET) biosensor is based on metal oxide field effect transistor that is gated by changes in the surfacecharges induced the reaction of biomolecules. In most cases of FET biosensor, FET biosensor is not being reused after thereaction; therefore, it is an important concept of investigate the biosensor with simplicity, cheap and reusability. However,the conventional cardiac troponin I (cTnI) sensing technique is inadequate owing to its low sensitivity and high operationaltime and cost. In this study, we developed a rapid and low-cost, and disposable electrical sensor using an extended gatefield-effect transistor (EGFET) to detect cTnI, as a key biomarker for myocardiac infarction. We first investigated pH sensingcharacteristics according to the pH level, which provided a logarithmically linear sensitivity in the pH sensing buffer solutionof approximately 57.9 mV/pH. Subsequently, we prepared a cTnI sample and monitored the reaction between cTnI andcTnI antibodies through the changes in the drain current and transfer curves. Our results showed that the EGFET biosensorcould successfully detect the cTnI levels as well as the pH with low-cost and rapid detection.
End-to-end Convolutional Neural Network Design for Automatic Detection of Influenza Virus
Junghwan Lee,Heesang Eom,Yuli Sun Hariyani,Cheonjung Kim,Yongkyoung Yoo,Jeonghoon Lee,Cheolsoo Park 대한전자공학회 2021 IEIE Transactions on Smart Processing & Computing Vol.10 No.1
Owing to the high mortality rate of influenza diseases, the early examination and accurate detection of the influenza virus are crucial for preventing potential tragedies. This paper reports the design of a highly reliable machine learning classifier for automatic detection of the influenza virus based on an image of its detection kit. Convolutional neural networks (CNNs), currently the most reliable image classifiers, were designed for the images of an influenza detection kit, and their hyperparameters were fine-tuned using an architecture search algorithm, Bayesian optimization, and hyperband (BOHB). With an overall accuracy of 90.14%, the designed and optimized 2DCNNs algorithm successfully separate the influenza virus from normal using the detection kit images.