• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.918-919
• /
• 2023

딥러닝 기술이 발전함에 따라 모델의 복잡성 역시 증가하고 있다. 본 연구에서는 모델 경량화를 위해 Knowledge Distillation 기법을 Crowd Counting Model에 적용했다. M-SFANet을 Teacher 모델로, 파라미터수가 적은 MCNN 모델을 Student 모델로 채택해 Knowledge Distillation을 적용한 결과, 기존의 MCNN 모델보다 성능을 향상했다. 이는 정확도와 메모리 효율성 측면에서 많은 개선을 이루어 컴퓨팅 리소스가 부족한 기기에서도 본 모델을 실행할 수 있어 많은 활용이 가능할 것이다.

• The Journal of Korea Robotics Society
• /
• v.10 no.2
• /
• pp.90-95
• /
• 2015

The main goal of e-learning systems is just-in-time knowledge acquisition. Rule-based e-learning systems, however, suffer from the mesa effect and the cold start problem, which both result in low user acceptance. E-learning systems suffer a further drawback in rendering the implementation of a natural interface in humanoids difficult. To address these concerns, even exceptional questions of the learner must be answerable. This paper aims to propose a method that can understand the learner's verbal cues and then intelligently explore additional domains of knowledge based on crowd data sources such as Wikipedia and social media, ultimately allowing for better answers in real-time. A prototype system was implemented using the NAO platform.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.541-542
• /
• 2021

In this paper, we consider the straggled network treatment in multiple sensor groups. During the moving or the implement of mission, the departed network should be handled. The loss of network group affects the mission implement of overall sensor networks, and it also requests the immediate treatment. In urgent mission environments, the network management of mobile sink should be considered to the group damage.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.7
• /
• pp.149-158
• /
• 2012

Much attention is being paid to social media because of their potentials to draw collective intelligence. In this context does this study attempt to draw some implications of social media as knowledge creation platform and suggest a conceptual framework of social media management. Information sharing among the public through social media literally produces profound influence throughout the society and thus not only business firms but all levels of public institutions, including government are seeking to take its advantage for various purposes such as public relations, crowd sourcing, etc. Especially considering that social media open the possibility of social knowledge creation platform in the Big Data era, this study is perhaps able to contribute to further development of social media management model together with a series of measuring indexes.

• Information Systems Review
• /
• v.16 no.2
• /
• pp.25-41
• /
• 2014

While previous studies on Crowd funding mainly focused on the plan and policy of the introduction of Crowd funding, this study classified Investment Decision Making criterion and proposed the significance of each factor, leading to results that can help investors practically. According to the analysis, (1) physical resource in business is the most important criterion to be considered as the way for investor to objectively make decisions on business model, (2) although people pay attention to the managerial capabilities of business, the creativity of business model is not being paid much attention. The result of this study can objectively determine investment decision making on business model thought a quantitative standard of judgment. Moreover, an environment should be built for healthy decision making which is based on business model instead of only product, and it is necessary to have institutional framework to enable new companies to have healthy competition with existing companies through creative business model.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.3
• /
• pp.156-162
• /
• 2013

Social manufacturing architecture is based on information delivery through social media and SNS, PLM and social platform, crowd source and knowledge base on whole phase of product life cycle. Although advanced social manufacturing issues are rising globally, its technology emerging into domestic industry is likely to be slow. We focused to review technological changes and R&D trend, patent issues, and then suggest assignments and advices to be practiced for social manufacturing modeling.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.23 no.2
• /
• pp.69-78
• /
• 2024

Recently, the use of CCTV to prevent crowd accidents has been promoted, but research is needed to compensate for the spatial limitations of CCTV. In this study, pedestrian density was measured using drone footage, and based on a review of existing literature, a threshold of 6.7 people/m² was selected as the cutoff risk level for crowd accidents. In addition, we conducted a preliminary study to determine drone parameters and found that the pedestrian recognition rate was high at a drone altitude of 20 meters and an angle of 60°. Based on a previous study, we selected a target area with a high concentration of pedestrians and measured pedestrian density, which was found to be 0.27~0.30 per m². The study shows it is possible to measure risk levels by determining pedestrian densities in target areas using drone images. We believe drone surveillance will be utilized for crowd safety management in the near future.

• Fire Science and Engineering
• /
• v.31 no.1
• /
• pp.36-41
• /
• 2017

In this study, the experimental measurements from a one-way moving experiment showed that the average movement speed was 0.55 m/s with an average crowd density of $2.36P/m^2$ in a corridor. The cCalculation result of the correlations between the crowd density and movement speed from the SFPE Handbook showed an average of 0.53 m/s. The difference between the calculation and experiment was 0.02 m/s. A comparison of each data set showed that the maximum difference was 0.38 m/s. Some experimental results showed that the crowd density increased with increasing movement speed and the average data from the entire experiment time was used for the analysis. When the short time interval for frame by frame analysis for video files was conducted, the experimental data was expected to be more reliable.

• Korean Journal of Construction Engineering and Management
• /
• v.19 no.1
• /
• pp.3-11
• /
• 2018

In order to find methods to encourage active private participation for efficient road maintenance, this study examined the status of the road improvement project and the development of the private capital utilization improvement project. Based on this, the study conducted a case study and a financial feasibility study that applied the new business method and selected the applicable business method. As a result, it is analyzed that the method of using Crowd Funding is more advantageous than the private scheme of the private capital utilization improvement project. This is because it will be highly effective in raising funds from users of highway users, especially users of improvement projects. In addition, it can also mean that the road users who suffer from inconvenience due to the construction can return the toll profit of road construction to the user.

• Asian Journal of Business Environment
• /
• v.10 no.4
• /
• pp.5-13
• /
• 2020

Purpose: The purpose of this paper is to empirically investigate the relationship between government subsidies and research and development (R&D) investment of animal husbandry companies in China. The moderating effects of firm size, debt ratio, and firm profitability on this relationship are also examined. Research design, data and methodology: The analysis is based on 14 animal husbandry companies listed on the Shanghai and Shenzhen stock exchanges over the period of 2012-2016. Data are obtained from the China Stock Market & Accounting Research (CSMAR) database and the RESSET database, and multiple regression analysis is utilized with the aid of Stata. Results: The empirical results show that government subsidies can promote R&D investment of animal husbandry companies in China. In addition, firm size, debt ratio, and firm profitability have positive moderating effects on the relationship between government subsidies and R&D investment. Conclusions: Based on the results, the paper concludes that government subsidies play an important role in the process of R&D of China's animal husbandry companies. This paper recommends that managers of animal husbandry companies should enhance the utilization efficiency of government subsidies and put great emphasis on R&D investment. The policymakers should implement more incentives to encourage animal husbandry companies to invest more in R&D.