• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.1
• /
• pp.1-10
• /
• 2022

Deep neural networks are widely used to solve various problems. In a fully connected neural network, the nonlinear activation function is a function that nonlinearly transforms the input value and outputs it. The nonlinear activation function plays an important role in solving the nonlinear problem, and various nonlinear activation functions have been studied. In this study, we propose a combined parametric activation function that can improve the performance of a fully connected neural network. Combined parametric activation functions can be created by simply adding parametric activation functions. The parametric activation function is a function that can be optimized in the direction of minimizing the loss function by applying a parameter that converts the scale and location of the activation function according to the input data. By combining the parametric activation functions, more diverse nonlinear intervals can be created, and the parameters of the parametric activation functions can be optimized in the direction of minimizing the loss function. The performance of the combined parametric activation function was tested through the MNIST classification problem and the Fashion MNIST classification problem, and as a result, it was confirmed that it has better performance than the existing nonlinear activation function and parametric activation function.

• Journal of Electrochemical Science and Technology
• /
• v.13 no.1
• /
• pp.148-158
• /
• 2022

The electrochemical CO₂ reduction (ECR) to produce value-added fuels and chemicals using clean energy sources (like solar and wind) is a promising technology to neutralize the carbon cycle and reproduce the fuels. Presently, the ECR has been the most attractive route to produce carbon-building blocks that have growing global production and high market demand. The electrochemical CO₂ reduction could be extensively implemented if it produces valuable products at those costs which are financially competitive with the present market prices. Herein, the electrochemical conversion of CO₂ obtained from flue gases of a power plant to produce diesel and formic acid using a consistent techno-economic approach is presented. The first scenario analyzed the production of diesel fuel which was formed through Fischer-Tropsch processing of CO (obtained through electroreduction of CO₂) and hydrogen, while in the second scenario, direct electrochemical CO₂ reduction to formic acid was considered. As per the base case assumptions extracted from the previous outstanding research studies, both processes weren't competitive with the existing fuel prices, indicating that high electrochemical (EC) cell capital cost was the main limiting component. The diesel fuel production was predicted as the best route for the cost-effective production of fuels under conceivable optimistic case assumptions, and the formic acid was found to be costly in terms of stored energy contents and has a facile production mechanism at those costs which are financially competitive with its bulk market price. In both processes, the liquid product cost was greatly affected by the parameters affecting the EC cell capital expenses, such as cost concerning the electrode area, faradaic efficiency, and current density.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.6_2
• /
• pp.729-735
• /
• 2021

Recently Artificial Intelligence(AI) has been developed and used in various fields. Especially AI recognition technology can perceive and distinguish images so it should plays a significant role in quality inspection process. For stability of autonomous driving technology, semiconductors inside automobiles must be protected from external electromagnetic wave(EM wave). As a shield film, a thin polymeric material with hole shaped micro-patterns created by a laser processing could be used for the protection. The shielding efficiency of the film can be increased by the hole structure with appropriate pitch and size. However, since the sensitivity of micro-machining for some parameters, the shape of every single hole can not be same, even it is possible to make defective patterns during process. And it is absolutely time consuming way to inspect all patterns by just using optical microscope. In this paper, we introduce a AI inspection system which is based on web site AI tool. And we evaluate the usefulness of AI model by calculate Area Under ROC curve(Receiver Operating Characteristics). The AI system can classify the micro-patterns into normal or abnormal ones displaying the text of the result on real-time images and save them as image files respectively. Furthermore, pressing the running button, the Hardware of robot arm with two Arduino motors move the film on the optical microscopy stage in order for raster scanning. So this AI system can inspect the entire micro-patterns of a film automatically. If our system could collect much more identified data, it is believed that this system should be a more precise and accurate process for the efficiency of the AI inspection. Also this one could be applied to image-based inspection process of other products.

• Journal of Internet Computing and Services
• /
• v.22 no.6
• /
• pp.115-127
• /
• 2021

With the development of science and technology around the world, the realm of cyberspace, following land, sea, air, and space, is also recognized as a battlefield area. Accordingly, it is necessary to design and establish various elements such as definitions, systems, procedures, and plans for not only physical operations in land, sea, air, and space but also cyber operations in cyberspace. In this research, the importance of cyber targets that can be considered when prioritizing the list of cyber targets selected through intermediate target development in the target development and prioritization stage of targeting processing of cyber operations was selected as a factor to be considered. We propose a method to calculate the score for the cyber target and use it as a part of the cyber target prioritization score. Accordingly, in the cyber target prioritization process, the cyber target importance category is set, and the cyber target importance concept and reference item are derived. We propose a TIR (Target Importance Rank) algorithm that synthesizes parameters such as Event Prioritization Framework based on PageRank algorithm for score calculation and synthesis for each derived standard item. And, by constructing the Stuxnet case-based network topology and scenario data, a cyber target importance score is derived with the proposed algorithm, and the cyber target is prioritized to verify the proposed algorithm.

• Journal of the Korea Convergence Society
• /
• v.13 no.3
• /
• pp.77-89
• /
• 2022

It has been proven through many previous researches that the pretrained language model with a large corpus helps improve performance in various natural language processing tasks. However, there is a limit to building a large-capacity corpus for training in a language environment where resources are scarce. Using the Cross-lingual Post-Training (XPT) method, we analyze the method's efficiency in Korean, which is a low resource language. XPT selectively reuses the English pretrained language model parameters, which is a high resource and uses an adaptation layer to learn the relationship between the two languages. This confirmed that only a small amount of the target language dataset in the relationship extraction shows better performance than the target pretrained language model. In addition, we analyze the characteristics of each model on the Korean language model and the Korean multilingual model disclosed by domestic and foreign researchers and companies.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.8
• /
• pp.371-380
• /
• 2023

Relation extraction is to extract relationships between named entities from text. Traditionally, relation extraction methods only extract relations between predetermined subject and object entities. However, in end-to-end relation extraction, all possible relations must be extracted by considering the positions of the subject and object for each pair of entities, and so this method uses time and resources inefficiently. To alleviate this problem, this paper proposes a method that sets directions based on the positions of the subject and object, and extracts relations according to the directions. The proposed method utilizes existing relation extraction data to generate direction labels indicating the direction in which the subject points to the object in the sentence, adds entity position tokens and entity type to sentences to predict the directions using a pre-trained language model (KLUE-RoBERTa-base, RoBERTa-base), and generates representations of subject and object entities through probabilistic crossover operation. Then, we make use of these representations to extract relations. Experimental results show that the proposed model performs about 3 ~ 4%p better than a method for predicting integrated labels. In addition, when learning Korean and English data using the proposed model, the performance was 1.7%p higher in English than in Korean due to the number of data and language disorder and the values of the parameters that produce the best performance were different. By excluding the number of directional cases, the proposed model can reduce the waste of resources in end-to-end relation extraction.

• Animal Bioscience
• /
• v.36 no.11
• /
• pp.1666-1684
• /
• 2023

Objective: Our objective was to examine the relationships of supplemental rumen-protected lysine (RPL) or lysine + methionine (RPLM) on lactational performance, plasma amino acids (AA) concentration, and nitrogen use efficiency of lactating dairy cows by using a meta-analysis approach. Methods: A total of 56 articles comprising 77 experiments with either RPL or RPLM supplementation were selected and analyzed using a mixed model methodology by considering the treatments and other potential covariates as fixed effects and different experiments as random effects. Results: In early lactating cows, milk yield was linearly increased by RPL (β₁ = 0.013; p<0.001) and RPLM (β₁ = 0.014; p<0.028) but 3.5% fat-corrected milk (FCM) and energy-corrected milk (ECM) (kg/d) was increased by only RPL. RPL and RPLM did not affect dry matter intake (DMI) but positively increased (p<0.05) dairy efficiency (Milk yield/DMI and ECM/DMI). As a percentage, milk fat, protein, and lactose were unchanged by RPL or RPLM but the yield of all components was increased (p<0.05) by feeding RPL while only milk protein was increased by feeding RPLM. Plasma Lys concentration was linearly increased (p<0.05) with increasing supplemental RPL while plasma Met increased (p<0.05) by RPLM supplementation. The increase in plasma Lys had a strong linear relationship (R² = 0.693 in the RPL dataset and R² = 0.769 in the RPLM dataset) on milk protein synthesis (g/d) during early lactation. Nitrogen metabolism parameters were not affected by feeding RPL or RPLM, either top-dress or when supplemented to deficient diets. Lactation performance did not differ between AA-deficient or AA-adequate diets in response to RPL or RPLM supplementation. Conclusion: RPL or RPLM showed a positive linear relationship on the lactational performance of dairy cows whereas greater improvement effects were observed during early lactation. Supplementing RPL or RPLM is recommended on deficient-AA diet but not on adequate-AA diet.

• Korean Journal of Radiology
• /
• v.22 no.3
• /
• pp.425-434
• /
• 2021

Objective: To investigate the potential value of ¹⁸F-fluorodeoxyglucose (FDG) PET/CT in predicting the survival of patients with primary tracheal malignant tumors. Materials and Methods: An analysis of FDG PET/CT findings in 37 primary tracheal malignant tumor patients with a median follow-up period of 43.2 months (range, 10.8-143.2 months) was performed. Cox proportional hazards regression analyses were used to assess the associations between quantitative ¹⁸F-FDG PET/CT parameters, other clinic-pathological factors, and overall survival (OS). A risk prognosis model was established according to the independent prognostic factors identified on multivariate analysis. A survival curve determined by the Kaplan-Meier method was used to assess whether the prognosis prediction model could effectively stratify patients with different risks factors. Results: The median survival time of the 37 patients with tracheal tumors was 38.0 months, with a 95% confidence interval of 10.8 to 65.2 months. The 3-year, 5-year and 10-year survival rate were 54.1%, 43.2%, and 16.2%, respectively. The metabolic tumor volume (MTV), total lesion glycolysis (TLG), maximum standardized uptake value, age, pathological type, extension categories, and lymph node stage were included in multivariate analyses. Multivariate analysis showed MTV (p = 0.011), TLG (p = 0.020), pathological type (p = 0.037), and extension categories (p = 0.038) were independent prognostic factors for OS. Additionally, assessment of the survival curve using the Kaplan-Meier method showed that our prognosis prediction model can effectively stratify patients with different risks factors (p < 0.001). Conclusion: This study shows that ¹⁸F-FDG PET/CT can predict the survival of patients with primary tracheal malignant tumors. Patients with an MTV > 5.19, a TLG > 16.94 on PET/CT scans, squamous cell carcinoma, and non-E1 were more likely to have a reduced OS.

• Journal of the Korean earth science society
• /
• v.37 no.7
• /
• pp.448-464
• /
• 2016

The purpose of this study is to optimize the parameters related to consensus coherency within the PCL 1300, the operating program of wind profiler, from a validation of wind data between rawinsonde and wind profiler at Chupungryeong ($36^{\circ}13^{\prime}$, $127^{\circ}59^{\prime}$) site in Korea. It is then to analyze the diurnal and seasonal characteristics of the turbulence energy dissipation rate (${\varepsilon}$) in clear and rainy days from March 2009 to February 2010. In comparison of the wind data between wind profiler and rawinsonde during April 22-23, 2010, it was shown in a big error more than $10ms^{-1}$ over the height of 3,000 meters in the zonal (u) and meridional (v) wind components. When removing more than $10ms^{-1}$ in each wind speed difference of u an v components between the two instruments, the correlation coefficients of these wind components were 0.92 and 0.88, respectively, and the root mean square errors were 3.07 and $1.06ms^{-1}$. Based on these results, when the data processing time and the minimum available data within the PCL 1300 program were adjusted as 30 minutes and 60%, respectively, the bias errors were small. In addition, as a result of an analysis of sensitivity to consensus coherency of u and v components within the PCL1300 program, u components were underestimated in radial coherency, instantaneous and winbarbs coherency, whereas v components were overestimated. Finally by optimizing parameters of the PCL1300 program, the diurnal and seasonal means of ${\varepsilon}$ at each height were higher in rainy days than those in clear days because of increasing in the vertical wind speed due to upward and downward motions. The mean ${\varepsilon}$ for clear and rainy days in winter was lower than those of other seasons, due to stronger horizontal wind speed in winter than those in other seasons. Consequently, when the turbulence energy dissipation rates in the vertical wind speed of more than ${\pm}10cm\;s^{-1}$ were excluded for clear and rainy days, the mean ${\varepsilon}$ in rainy days was 6-7 times higher than that in clear days, but when considering them, it was 4-5 times higher.

• KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
• /
• v.22 no.3
• /
• pp.95-102
• /
• 2016

This study investigated the effects of the oxygen permeability of vacuum packaging film and the storage temperature on the quality and shelf life of Hamburg steaks during storage for 14 days. Control samples (C) were packaged in a polyamide/polyethylene (PA/PE) film and stored at $5^{\circ}C$. Treatment samples were either packaged in an ethylene vinyl alcohol/polyethylene (EVOH/PE) copolymer film and stored at $5^{\circ}C$ (T1), and in a PA/PE film and stored at $-18^{\circ}C$ (T2). The initial total plate count (TPC) was 3.6 log cfu/g. In T1 samples, TPC and Brochothrix thermosphacta counts were increased, similar to those in C samples, whereas Pseudomonas spp. counts were significantly lower than those in C samples during storage. Over the storage period, the volatile basic nitrogen values increased most rapidly in C samples, followed by T1 and then T2 samples. The values of thiobarbituric acid reactive substances steadily increased in all samples during storage. The colour parameters were not significantly different among the samples during storage. T1 samples maintained sensory qualities in flavour and off-odour parameters for two days longer than C samples did. At day 12, T2 samples were evaluated as being below the marketability score of 5.0 for texture. In conclusion, using high oxygen barrier films like EVOH/PE copolymer for packaging Hamburg steaks could extend the sensory qualities in view of flavour and off-odour during chilled storage. However, frozen storage at $-18^{\circ}C$ is recommended when the storage period is extended beyond 14 days at $5^{\circ}C$.