• Steel and Composite Structures
• /
• v.51 no.4
• /
• pp.441-456
• /
• 2024

This paper aims to develop Machine Learning (ML) algorithms to predict the buckling resistance of cold-formed steel (CFS) channels with restrained flanges, widely used in typical CFS sheathed wall panels, and provide practical design tools for engineers. The effects of cross-sectional restraints were first evaluated on the elastic buckling behaviour of CFS channels subjected to pure axial compressive load or bending moment. Feedforward multi-layer Artificial Neural Networks (ANNs) were then trained on different datasets comprising CFS channels with various dimensions and properties, plate thicknesses, and restraining conditions on one or two flanges, while the elastic distortional buckling resistance of the elements were determined according to the Finite Strip Method (FSM). To develop less biased networks and ensure that every observation from the original dataset has the chance of appearing in the training and test set, a K-fold cross-validation technique was implemented. In addition, the hyperparameters of the ANNs were tuned using a grid search technique to provide ANNs with optimum performances. The results demonstrated that the trained ANNs were able to predict the elastic distortional buckling resistance of CFS flange-restrained elements with an average accuracy of 99% in terms of coefficient of determination. The developed models were then used to propose a simple ANN-based design formula for the prediction of the elastic distortional buckling stress of CFS flange-restrained elements. Finally, the proposed formula was further evaluated on a separate set of unseen data to ensure its accuracy for practical applications.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.7-21
• /
• 2024

There are various machine learning techniques such as Reinforcement Learning, Deep Learning, Neural Network Learning, and so on. In recent, Large Language Models (LLMs) are popularly used for Generative AI based on Reinforcement Learning. It makes decisions with the most optimal rewards through the fine tuning process in a particular situation. Unfortunately, LLMs can not provide any explanation for how they reach the goal because the training is based on learning of black-box AI. Reinforcement Learning as black-box AI is based on graph-evolving structure for deriving enhanced solution through adjustment by human feedback or reinforced data. In this research, for mutually exclusive decision-making, Mutually Exclusive Learning (MEL) is proposed to provide explanations of the chosen goals that are achieved by a decision on both ends with specified conditions. In MEL, decision-making process is based on the tree-based structure that can provide processes of pruning branches that are used as explanations of how to achieve the goals. The goal can be reached by trade-off among mutually exclusive alternatives according to the specific contextual conditions. Therefore, the tree-based structure is adopted to provide feasible solutions with the explanations based on the pruning branches. The sequence of pruning processes can be used to provide the explanations of the inferences and ways to reach the goals, as Explainable AI (XAI). The learning process is based on the pruning branches according to the multi-dimensional contextual conditions. To deep-dive the search, they are composed of time window to determine the temporal perspective, depth of phases for lookahead and decision criteria to prune branches. The goal depends on the policy of the pruning branches, which can be dynamically changed by configured situation with the specific multi-dimensional contextual conditions at a particular moment. The explanation is represented by the chosen episode among the decision alternatives according to configured situations. In this research, MEL adopts the tree-based learning model to provide explanation for the goal derived with specific conditions. Therefore, as an example of mutually exclusive problems, employment process is proposed to demonstrate the decision-making process of how to reach the goal and explanation by the pruning branches. Finally, further study is discussed to verify the effectiveness of MEL with experiments.

• IMMUNE NETWORK
• /
• v.24 no.4
• /
• pp.32.1-32.13
• /
• 2024

Low-dose radiotherapy (LDRT) has been explored as a treatment option for various inflammatory diseases; however, its application in the context of rheumatoid arthritis (RA) is lacking. This study aimed to elucidate the mechanism underlying LDRT-based treatment for RA and standardize it. LDRT reduced the total numbers of immune cells, but increased the apoptotic CD4⁺ T and B220⁺ B cells, in the draining lymph nodes of collagen induced arthritis and K/BxN models. In addition, it significantly reduced the severity of various pathological manifestations, including bone destruction, cartilage erosion, and swelling of hind limb ankle. Post-LDRT, the proportion of apoptotic CD4⁺ T and CD19⁺ B cells increased significantly in the PBMCs derived from human patients with RA. LDRT showed a similar effect in fibroblast-like synoviocytes as well. In conclusion, we report that LDRT induces apoptosis in immune cells and fibro-blast-like synoviocytes, contributing to attenuation of arthritis.

• Journal of IKEEE
• /
• v.28 no.3
• /
• pp.352-359
• /
• 2024

This paper proposes a CNN-ViT hybrid model that automatically evaluates the aesthetic quality of images by combining local and global features. In this approach, CNN is used to extract local features such as color and object placement, while ViT is employed to analyze the aesthetic value of the image by reflecting global features. Color composition is derived by extracting the primary colors from the input image, creating a color palette, and then passing it through the CNN. The Rule of Thirds is quantified by calculating how closely objects in the image are positioned near the thirds intersection points. These values provide the model with critical information about the color balance and spatial harmony of the image. The model then analyzes the relationship between these factors to predict scores that align closely with human judgment. Experimental results on the AADB image database show that the proposed model achieved a Spearman's Rank Correlation Coefficient (SRCC) of 0.716, indicating more consistent rank predictions, and a Pearson Correlation Coefficient (LCC) of 0.72, which is 2~4% higher than existing models.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.5
• /
• pp.1021-1035
• /
• 2024

This paper proposes the TEE-BOP (Trusted Execution Environment-Based Blockchain Offline Payment) protocol for blockchain-based offline payments. TEE-BOP securely manages offline balances within a Trusted Execution Environment (TEE) and efficiently verifies initial deposit proofs recorded on the blockchain using Merkle trees. Additionally, it ensures secure and tamper-proof transactions in offline environments by guaranteeing the reliability of keys and the system through TEE Attestation. Unlike previous studies, TEE-BOP enhances real-world applicability by eliminating dependence on central authorities and avoiding assumptions of ideal models. The protocol solves the double-spending problem through multi-layered defense mechanisms and addresses forgery prevention by allowing recipients to directly verify data consistency between the TEE and the blockchain. This enables reliable blockchain-based offline payments in areas with unstable network infrastructure. It demonstrates that this research can expand the application of blockchain technology and contribute to improving access to financial services in developing countries or disaster situations.

• Geophysics and Geophysical Exploration
• /
• v.13 no.1
• /
• pp.80-87
• /
• 2010

This study was conducted to develop a reservoir modelling workflow to reproduce the heterogeneous distribution of effective permeability that impacts on the performance of SAGD (Steam Assisted Gravity Drainage), the in-situ bitumen recovery technique in the Athabasca Oil Sands. Lithologic facies distribution is the main cause of the heterogeneity in bitumen reservoirs in the study area. The target formation consists of sand with mudstone facies in a fluvial-to-estuary channel system, where the mudstone interrupts fluid flow and reduces effective permeability. In this study, the lithologic facies is classified into three classes having different characteristics of effective permeability, depending on the shapes of mudstones. The reservoir modelling workflow of this study consists of two main modules; facies modelling and permeability modelling. The facies modelling provides an identification of the three lithologic facies, using a stochastic approach, which mainly control the effective permeability. The permeability modelling populates mudstone volume fraction first, then transforms it into effective permeability. A series of flow simulations applied to mini-models of the lithologic facies obtains the transformation functions of the mudstone volume fraction into the effective permeability. Seismic data contribute to the facies modelling via providing prior probability of facies, which is incorporated in the facies models by geostatistical techniques. In particular, this study employs a probabilistic neural network utilising multiple seismic attributes in facies prediction that improves the prior probability of facies. The result of using the improved prior probability in facies modelling is compared to the conventional method using a single seismic attribute to demonstrate the improvement in the facies discrimination. Using P-wave velocity in combination with density in the multiple seismic attributes is the essence of the improved facies discrimination. This paper also discusses sand matrix porosity that makes P-wave velocity differ between the different facies in the study area, where the sand matrix porosity is uniquely evaluated using log-derived porosity, P-wave velocity and photographically-predicted mudstone volume.

• Journal of Korean Society of Transportation
• /
• v.22 no.3 s.74
• /
• pp.145-157
• /
• 2004

The cycle length design model of the Korean traffic responsive signal control systems is devised to vary a cycle length as a response to changes in traffic demand in real time by utilizing parameters specified by a system operator and such field information as degrees of saturation of through phases. Since no explicit guideline is provided to a system operator, the system tends to include ambiguity in terms of the system optimization. In addition, the cycle lengths produced by the existing model have yet been verified if they are comparable to the ones minimizing delay. This paper presents the studies conducted (1) to find shortcomings embedded in the existing model by comparing the cycle lengths produced by the model against the ones minimizing delay and (2) to propose a new direction to design a cycle length minimizing delay and excluding such operator oriented parameters. It was found from the study that the cycle lengths from the existing model fail to minimize delay and promote intersection operational conditions to be unsatisfied when traffic volume is low, due to the feature of the changed target operational volume-to-capacity ratio embedded in the model. The 64 different neural network based cycle length design models were developed based on simulation data surrogating field data. The CORSIM optimal cycle lengths minimizing delay were found through the COST software developed for the study. COST searches for the CORSIM optimal cycle length minimizing delay with a heuristic searching method, a hybrid genetic algorithm. Among 64 models, the best one producing cycle lengths close enough to the optimal was selected through statistical tests. It was found from the verification test that the best model designs a cycle length as similar pattern to the ones minimizing delay. The cycle lengths from the proposed model are comparable to the ones from TRANSYT-7F.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.10
• /
• pp.303-308
• /
• 2020

The ReLU(Rectified Linear Unit) function has been dominantly used as a standard activation function in most deep artificial neural network models since it was proposed. Later, Leaky ReLU, Swish, and Mish activation functions were presented to replace ReLU, which showed improved performance over existing ReLU function in image classification task. Therefore, we recognized the need to experiment with whether performance improvements could be achieved by replacing the RELU with other activation functions in the super resolution task. In this paper, the performance was compared by changing the activation functions in EDSR model, which showed stable performance in the super resolution task. As a result, in experiments conducted with changing the activation function of EDSR, when the resolution was converted to double, the existing activation function, ReLU, showed similar or higher performance than the other activation functions used in the experiment. When the resolution was converted to four times, Leaky ReLU and Swish function showed slightly improved performance over ReLU. PSNR and SSIM, which can quantitatively evaluate the quality of images, were able to identify average performance improvements of 0.06%, 0.05% when using Leaky ReLU, and average performance improvements of 0.06% and 0.03% when using Swish. When the resolution is converted to eight times, the Mish function shows a slight average performance improvement over the ReLU. Using Mish, PSNR and SSIM were able to identify an average of 0.06% and 0.02% performance improvement over the RELU. In conclusion, Leaky ReLU and Swish showed improved performance compared to ReLU for super resolution that converts resolution four times and Mish showed improved performance compared to ReLU for super resolution that converts resolution eight times. In future study, we should conduct comparative experiments to replace activation functions with Leaky ReLU, Swish and Mish to improve performance in other super resolution models.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.49 no.2
• /
• pp.61-73
• /
• 2021

This study aimed to identify the landscape perception and landscape characteristics of users by utilizing SNS data generated by their experiences. Therefore, how to recognize the main places and scenery appearing on the island, and what are the characteristics of the main scenery were analyzed using online text data and photo data. Text data are text mining and network structural analysis, while photographic data are landscape identification models and color analysis. As a result of the study, First, as a result of frequency analysis of Bakji·Banwol-do topics, we were able to derive keywords for local landscapes such as 'Purple Bridge', 'Doori Village', and location, behavior, and landscape images by analyzing them simultaneously. Second, the network structure analysis showed that the connection between key and undrawn keywords could be more specifically analyzed, indicating that creating landscapes using colors is affecting regional activation. Third, after analyzing the landscape identification model, it was found that artificial elements would be excluded to create preferred landscapes using the main targets of "Purple Bridge" and "Doori Village", and that it would be effective to set a view point of the sea and sky. Fourth, Bakji·Banwol-do were the first islands to be created under the theme of color, and the colors used in artificial facilities were similar to the surrounding environment, and were harmonized with contrasting lighting and saturation values. This study used online data uploaded directly by visitors in the landscape field to identify users' perceptions and objects of the landscape. Furthermore, the use of both text and photographic data to identify landscape recognition and characteristics is significant in that they can specifically identify which landscape and resources they prefer and perceive. In addition, the use of quantitative big data analysis and qualitative landscape identification models in identifying visitors' perceptions of local landscapes will help them understand the landscape more specifically through discussions based on results.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.17 no.1
• /
• pp.121-137
• /
• 2022

The purpose of this study is to present the necessary success factors and strategies for high-tech social ventures and stakeholders in the related ecosystem by empirically identifying factors that affect their sustainable performance. Based on prior research, the dimensions of three performance factors were presented: core technology competency, core business competency, and social mission orientation. Then, such sub-dimensions such as technology innovation orientation, R&D capability, business model, customer orientation, social network, and social mission pursuit were derived. For empirical analysis, a survey was conducted on domestic high-tech social ventures, and the significance of the hypothesis was tested through PLS-structural equation analysis of the collected 243 valid data. As a result, it was found that the technology innovation orientation was embedded as an abstract organizational and cultural characteristic in the high-tech social venture, which is a research sample, and thus did not significantly affect the dependent variable. In other words, aiming for the latest cutting-edge technology alone cannot affect performance, and it is a result of proving the need for substantial influencing factors that can strengthen it. On the other hand, the business model had a significant effect only on social performance, which is presumed to be the limitation of measurement tools developed for social enterprises, and the results of additional multi-group analysis to determine the cause also supported the basis for this estimation. Excluding the previous two performance factors, R&D competency, customer orientation, social network, and social mission pursuit were all found to have a significant positive (+) effect on social and economic performance. This study laid a foundation for related research by identifying high-tech social ventures emerging in the ecosystem of a social economy and expanded empirical research models related to the performance of existing social enterprises and social ventures. However, in the research method or process, there were limitations such as factor derivation or verification for balance of dual performance, subjective measurement method, and sample representativeness. It is expected that more in-depth follow-up studies will continue by supplementing future limitations and designing improved research models.