http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
SCM에서 흡수역량이 조직성과, 그리고 기업성과에 미치는 영양요인에 대한 연구
조근식(Jo, Geunsik),구철모(Goo, Chulmo) 인하대학교 정석물류통상연구원 2008 인하대학교 정석물류통상연구원 연구총서 Vol.- No.-
Many companies demand capability of acquiring external knowledge and internalizing it to add their knowledge capability to survive in competitive environment, The paper analyzes the impact of absorptive acpacity, an ability to recognize and utilize external knowledge, on knowledge acquisition of the firm, and examines the role of absorptive capacity in the supply chain context. Data were collected from 200 domestic companies that had adopted SCM. The results indicate that prior experiences of knowledge acquisition, transportation, transformation, and exploitation affect knowledge acquistion positively. Among those independent variables, absorptive mechanism significantly and positively related to the firm performance. Knowledge creation, organizational performance are overally related. Further, the results provide evidence that SCM plays a independent roel for knowledge acquisition.
Morpho-GAN: Unsupervised Learning of Data with High Morphology using Generative Adversarial Networks
Azamat Abduazimov(아자맛 압두아지모프),GeunSik Jo(조근식) 한국컴퓨터정보학회 2020 한국컴퓨터정보학회 학술발표논문집 Vol.28 No.1
The importance of data in the development of deep learning is very high. Data with high morphological features are usually utilized in the domains where careful lens calibrations are needed by a human to capture those data. Synthesis of high morphological data for that domain can be a great asset to improve the classification accuracy of systems in the field. Unsupervised learning can be employed for this task. Generating photo-realistic objects of interest has been massively studied after Generative Adversarial Network (GAN) was introduced. In this paper, we propose Morpho-GAN, a method that unifies several GAN techniques to generate quality data of high morphology. Our method introduces a new suitable training objective in the discriminator of GAN to synthesize images that follow the distribution of the original dataset. The results demonstrate that the proposed method can generate plausible data as good as other modern baseline models while taking a less complex during training.
적대적 생성 신경망을 이용한 얼굴 감정인식 데이터 증강
김진용(Jinyong Kim),조근식(Geunsik Jo) Korean Institute of Information Scientists and Eng 2021 정보과학회논문지 Vol.48 No.4
The facial emotion recognition field of computer vision has recently been identified to demonstrate meaningful results through various neural networks. However, the major datasets of facial emotion recognition have the problem of “class imbalance,” which is a factor that degrades the accuracy of deep learning models. Therefore, numerous studies have been actively conducted to solve the problem of class imbalance. In this paper, we propose “RDGAN,” a facial emotion recognition data augmentation model that uses a GAN to solve the class imbalance of the FER2013 and RAF_single that are used as facial emotion recognition datasets. RDGAN is a network that generates images suitable for classes by adding expression discriminators based on the image-to-image translation model between the existing images as compared to the prevailing studies. The dataset augmented with RDGAN showed an average performance improvement of 4.805%p and 0.857%p in FER2013 and RAF_single, respectively, compared to the dataset without data augmentation.
C3D와 객체 기반의 움직임 정보 결합을 통한 감시시스템에서의 이상 행동 탐지
박슬기(Seulgi Park),홍명덕(Myungduk Hong),조근식(Geunsik Jo) Korean Institute of Information Scientists and Eng 2021 정보과학회논문지 Vol.48 No.1
In the existing closed-circuit television (CCTV) videos, the deep learning-based anomaly detection reported in the literature detected anomalies using only the objects action value. For this reason, it is difficult to extract the action value of an object depending upon the situation, and there is a problem that information is reduced over time. Since the cause of abnormalities in CCTV videos involves several factors such as frame complexity and information according to time series analysis, there is a limit to detecting an abnormality using only the action value of the object. To solve this problem, in this paper, we designed a new deep learning-based anomaly detection model that combined optical flow with C3D to use various feature values centered on the objects. The proposed anomaly detection model used the UCF-Crime dataset, and the experimental results achieved an area under the curve (AUC) of 76.44. Compared to previous studies, this study worked more effectively in fast-moving videos such as explosions. Finally, we concluded that it was appropriate to use the information according to different feature values and time series analysis considering various aspects of the behavior of an object when designing an anomaly detection model.
정진국(JinGuk Jung),이순근(SoonGeun Lee),조근식(GeunSik Jo) 한국지능정보시스템학회 2003 지능정보연구 Vol.9 No.3
In context of e-commerce, negotiation is a procedure to help negotiate between buyer and seller by adjusting their negotiation issues such as price and in terms of payment. We used intelligent agent and mobile device to promote new framework of e-commerce. Moreover, this framework can help buyers and sellers to carry their commercial transactions effectively. In regard to that issue, we need to carry out the research of negotiation agent that can be used in e-commerce fields.<br/> In this paper, we modeled the negotiation using CSP for the performance of agent in m-commerce environment. Furthermore we implemented interface for mobile device to extract buyer's requirement and preference easily. Besides that we used utility function to make a decision for various evaluation functions and suggestions that are used for evaluation of negotiation issues. A difficulty of generating offer is dependent on the number of negotiation issues and the range of the values. Therefore, if any offer has a number of negotiation issues and the range of values are wide, the search space will be exponentially expanded. There have been many studies for solving this problem, we applied those techniques to improve the agent's ability of negotiation. For example, a contract can be accomplished by exchanging seller and buyer's offer that is generated by agent to adjust the requisite profit for each party. Finally, we show the improvement of satisfaction as the negotiation is processed.
스케줄링 문제 해결을 위한 지식 기반 기법과 제약 만족 기법의 비교 연구
양종윤(Jongyoon Yang),정종진(JongJin Jung),조근식(Geunsik Jo) 한국정보과학회 1997 한국정보과학회 학술발표논문집 Vol.24 No.2Ⅱ
다양한 산업영역에서 수행되는 스케줄링 문제를 해결하기 위하여 AI분야에서는 지식을 기반으로 한 방법이 적용되어 왔다. 그러나 최근 CSP(Constraints Satisfaction Problem) 개념이 소개되어 그 효율성이 입증되고 있으며 스케줄링 응용 문제들이 CSP로 정형화되면서부터 지식 기반 기법과 제약만족 기법의 적용이 공존하고 있다. 지식을 기반으로 한 방법은 도메인 전문가(domain expert)의 지식을 습득하여 시스템에 반영하는데 이러한 지식은 문제해결에 중심적 역할을 수행하게 된다. 제약 조건을 기반으로 한 방법은 문제를 CSP로 정형화 한 후 제약조건에 따른 일관성 유지 및 휴리스틱 탐색 방법을 적용하여 문제의 해를 효율적으로 구하게 된다. 본 연구에서는 스케줄링 문제를 해결하기 위한 지식기반 기법과 제약만족 기법을 주기장 할당 문제에 적용하여 실제 항공사의 운항 데이터를 바탕으로 실험하고 분석 및 비교를 통해 제약 만족 기법이 시스템의 유지 및 보수 측면에서 효율적이며 근사해가 아닌 최적해를 통한 문제 해결이 가능함을 보였다.
Backbone Network for Object Detection with Multiple Dilated Convolutions and Feature Summation
Vani Natalia Kuntjono(바니 나탈리아 쿤트조노),Seunghyun Ko(고승현),Yang Fang(방양),Geunsik Jo(조근식) Korean Institute of Information Scientists and Eng 2018 정보과학회논문지 Vol.45 No.8
The advancement of CNN leads to the trend of using very deep convolutional neural network which contains more than 100 layers not only for object detection, but also for image segmentation and object classification. However, deep CNN requires lots of resources, and so is not suitable for people who have limited resources or real time requirements. In this paper, we propose a new backbone network for object detection with multiple dilated convolutions and feature summation. Feature summation enables easier flow of gradients and minimizes loss of spatial information that is caused by convolving. By using multiple dilated convolution, we can widen the receptive field of individual neurons without adding more parameters. Furthermore, by using a shallow neural network as a backbone network, our network can be trained and used in an environment with limited resources and without pre-training it in ImageNet dataset. Experiments demonstrate we achieved 71% and 38.2% of accuracy on Pascal VOC and MS COCO dataset, respectively.
이현진(Hyunjin Lee),김원택(Yuanze Jin),유영훈(Younghoon Yu),조근식(Geunsik Jo) 한국정보과학회 2009 한국정보과학회 학술발표논문집 Vol.36 No.1
차량경로문제는 지리적으로 산재해 있는 고객들에게 서비스를 수행하고 다시 차고지로 복귀하는데 소요되는 비용을 최소화하도록 차량의 경로를 결정하는 문제이다. 본 논문에서는 NP-hard 문제로 잘 알려진 차량경로문제를 풀기 위하여 빈발 경로 패턴을 이용한 메타 휴리스틱 해법을 제안한다. 이는 좋은 해의 집합에서 빈번히 발생하는 경로의 패턴을 이용하여 더욱 합리적인 해의 공간을 탐색함으로써 개선된 해를 발견할 수 있게 해준다. 본 메타 휴리스틱의 성능 비교를 위하여 기존의 알려진 다양한 휴리스틱들과 비교 실험을 하였으며 기존의 결과보다 향상되었음을 보였다.