• 한국인공지능학회지
• /
• 제7권2호
• /
• pp.9-12
• /
• 2019

With the Chinese government's attention to the artificial intelligence industry, the Chinese government has invested a lot in it recently. Of course, the importance of artificial intelligence industry for China's economic development is increasingly significant. The advent of artificial intelligence boom has also triggered a large number of scientists to analyze the impact of artificial intelligence on economic growth. Therefore, this paper use 31 China's cross-province panel data to study the effect of artificial intelligence on economic growth. Via empirical analyses under a series of econometric methods such as the province and year fixed effect model, the empirical result shows that artificial intelligence has a positive and significant effect on economic growth. Namely, the artificial intelligence is a new engine for economic growth. Meanwhile, the empirical results also indicate that the investment and consumption has a significant and positive effect on economic growth. Oppositely, the inflation and government purchase have a significant negative effect on economic growth. These findings in this paper also provide some important evidences for policy-makers to perform precise behaviors so as to promote the economic growth. Moreover, these finding enriches existing literature on artificial intelligence and economic growth.

• 한국중재학회지:중재연구
• /
• 제29권3호
• /
• pp.99-119
• /
• 2019

In recent times, the volume of cross-border e-commerce has witnessed an upward trend and has been accompanied by increased disputes, with cross-border e-commerce being characterized mainly by low value and large volume issues. For this reason, Online Dispute Resolution (ODR) was formed to carry out dispute resolutions in cross-border e-commerce. A virtual multi-door ODR platform for small value, cross-border disputes in e-commerce is then proposed in this paper. For a couple of decades, researchers have tried to employ Artificial Intelligence (AI) to Law. However, it turns out that they were faced with a couple of obstacles to integrate AI to Law since it is highly difficult to program AI to process the common sense of a human being. For example, AI cannot assimilate the affective side of a human being, and it is problematic to integrate a human being's common sense into the AI system. Considering this situation, this study puts forward an ODR model for cross-border e-commerce in the evolutionary perspective.

• 한국염색가공학회지
• /
• 제34권4호
• /
• pp.272-283
• /
• 2022

In this study, the flat-section monofilaments of PBT for artificial eyelashes was developed, and the physical properties of the circular cross-section of artificial eyelashes were compared and observed, and the main performance of the artificial eyelash prototype was observed through processing for artificial eyelashes. In addition, a satisfaction survey of the prototype was conducted through a survey of consumers and artificial eyelash operators. It was found that the bending stiffness value of the monofilaments increased significantly as the thickness increased. As a result of measuring the bending properties of the flat-section PBT monofilaments, the bending stiffness was significantly lower than that of the circular-section PBT specimens of the same thickness. The deformed cross-section PBT monofilaments with flat cross sections developed in this study showed a light weight factor of less than 50% compared to the existing circular cross-section PBT ones. The adhesive strength of the developed PBT artificial specimens was greater than that of the existing circular cross-section yarn. It was also observed that the curl stability over time was excellent. As a result of the consumer survey, it was possible to obtain more than 85% of positive answers in the case of consumer subjects, and it was possible to investigate that the satisfaction of the operator subjects was more than 80% compared to the existing round-section eyelashes.

• Biomaterials and Biomechanics in Bioengineering
• /
• 제1권3호
• /
• pp.163-174
• /
• 2014

The objective of this study is to prepare an artificial extracellular matrix (ECM) for cell culture by using polymer hydrogels. The polymer used is a cytocompatible water-soluble phospholipid polymer: poly[2-methacryloyloxyethyl phosphorylcholine (MPC)-n-butyl methacrylate-p-nitrophenyloxycarbonyl poly(ethylene oxide) methacrylate (MEONP)] (PMBN). The hydrogels are prepared using a cross-linking reaction between PMBN and diamine compounds, which can easily react to the MEONP moiety under mild conditions. The most favorable diamine is the bis(3-aminopropyl) poly(ethylene oxide) (APEO). The effects of cross-linking density and the chemical structure of cross-linking molecules on the mechanical properties of the hydrogel are evaluated. The storage modulus of the hydrogel is tailored by tuning the PMBN concentration and the MEONP/amino group ratio. The porous structure of the hydrogel networks depends not only on these parameters but also on the reaction temperature. We prepare a hydrogel with $40-50{\mu}m$ diameter pores and more than 90 wt% swelling. The permeation of proteins through the hydrogel increases dramatically with an increase in pore size. To induce cell adhesion, the cell-attaching oligopeptide, RGDS, is immobilized onto the hydrogel using MEONP residue. Bovine pulmonary artery endothelial cells (BPAECs) are cultured on the hydrogel matrix and are able to migrate into the artificial matrix. Hence, the RGDS-modified PMBN hydrogel matrix with cross-linked APEO functions as an artificial ECM for growing cells for applications in tissue engineering.

• 한국수산과학회지
• /
• 제41권3호
• /
• pp.215-220
• /
• 2008

This study examined the stability of cross- and box-type artificial reefs used to construct seaweed beds. Theoretical and experimental studies have clarified the Froude similitude. However, basic data needed to develop a more stable design for artificial reefs must be collected from long-term studies and analyses of sliding caused by waves. Hydraulic experiments are important for solving problems in the design and construction of artificial reefs. This study examined some design parameters for artificial reefs under wave and currents. The results showed the stability of cross- and box-type artificial reefs for constructing seaweed beds using a dimensionless parameter (the surf similarity parameter), water particle velocity, and so on. The hydraulics experiment indicated that the stability of artificial reefs differed according to their method of installation. This implies that artificial reefs should be installed after considering various environmental factors, such as wave breaking, reflection, and sediments.

• 한국인공지능학회지
• /
• 제5권1호
• /
• pp.1-9
• /
• 2017

Recently, the Q-Learning algorithm, which is one kind of reinforcement learning, is mainly used to implement artificial intelligence system in combination with deep learning. Many research is going on to improve the performance of Q-Learning. Therefore, purpose of theory try to improve the performance of Q-Learning algorithm. This Theory apply Cross Entropy Error to the loss function of Q-Learning algorithm. Since the mean squared error used in Q-Learning is difficult to measure the exact error rate, the Cross Entropy Error, known to be highly accurate, is applied to the loss function. Experimental results show that the success rate of the Mean Squared Error used in the existing reinforcement learning was about 12% and the Cross Entropy Error used in the deep learning was about 36%. The success rate was shown.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.298-301
• /
• 2004

The aim of this study is to cross-validate of spatial probability model, artificial neural network at Boun, Korea, using a Geographic Information System (GIS). Landslide locations were identified in the Boun, Janghung and Youngin areas from interpretation of aerial photographs, field surveys, and maps of the topography, soil type, forest cover and land use were constructed to spatial data-sets. The factors that influence landslide occurrence, such as slope, aspect and curvature of topography, were calculated from the topographic database. Topographic type, texture, material, drainage and effective soil thickness were extracted from the soil database, and type, diameter, age and density of forest were extracted from the forest database. Lithology was extracted from the geological database, and land use was classified from the Landsat TM image satellite image. Landslide susceptibility was analyzed using the landslide­occurrence factors by artificial neural network model. For the validation and cross-validation, the result of the analysis was applied to each study areas. The validation and cross-validate results showed satisfactory agreement between the susceptibility map and the existing data on landslide locations.

• Journal of Information Processing Systems
• /
• 제18권4호
• /
• pp.535-548
• /
• 2022

Using a rich resource language to classify sentiments in a language with few resources is a popular subject of research in natural language processing. Burmese is a low-resource language. In light of the scarcity of labeled training data for sentiment classification in Burmese, in this study, we propose a method of transfer learning for sentiment analysis of a language that uses the feature transfer technique on sentiments in English. This method generates a cross-language word-embedding representation of Burmese vocabulary to map Burmese text to the semantic space of English text. A model to classify sentiments in English is then pre-trained using a convolutional neural network and an attention mechanism, where the network shares the model for sentiment analysis of English. The parameters of the network layer are used to learn the cross-language features of the sentiments, which are then transferred to the model to classify sentiments in Burmese. Finally, the model was tuned using the labeled Burmese data. The results of the experiments show that the proposed method can significantly improve the classification of sentiments in Burmese compared to a model trained using only a Burmese corpus.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2024년도 춘계학술발표대회
• /
• pp.709-710
• /
• 2024

Super resolution에서 High-frequency Details를 개선하는 것이 최근 문제이다. 기존에는 Super resolution을 Regression task로 접근하므로써 L2 Loss를 사용하여 이미지가 흐릿하게 되었다. 이를 해결하기위해, Classification task로 바꾸므로써 Cross Entropy Loss을 적용하여 Cross-entropy Super-resolution (CS)를 설계한다. CS를 통해 선명도와 Details이 개선되지만, 저주파의 CE Loss 학습으로인한 Black Artifacts가 발생한다. 그래서, L2 Loss는 저주파와 같이 큰 신호에 더 초점을 맞추므로, 성능 개선을 위해 저주파를 L2 Loss에서, 고주파를 CE Loss에서 학습시킨 Frequency-specific Cross-entropy Super-resolution (FCS)을 제안한다. 우리는 왜곡에 강하며 Human의 인식과 유사한 측정지표인 Learned Perceptual Image Patch Similarity (LPIPS)로 평가한다. 실험한 모든 데이터 셋에서 우리의 FCS는 Baseline보다 LPIPS가 약 1.7배 정도 개선되었다.