• Korean Chemical Engineering Research
• /
• v.61 no.3
• /
• pp.388-393
• /
• 2023

The analysis of engineering data using neural network based on supervised learning has been utilized in various engineering fields such as optimization of chemical engineering process, concentration prediction of particulate matter pollution, prediction of thermodynamic phase equilibria, and prediction of physical properties for transport phenomena system. The supervised learning requires training data, and the performance of the supervised learning is affected by the composition and the configurations of the given training data. Among the frequently observed engineering data, the data is given in log-scale such as length of DNA, concentration of analytes, etc. In this study, for widely distributed log-scaled training data of virtual 100×100 images, available loss functions were quantitatively evaluated in terms of (i) confusion matrix, (ii) maximum relative error and (iii) mean relative error. As a result, the loss functions of mean-absolute-percentage-error and mean-squared-logarithmic-error were the optimal functions for the log-scaled training data. Furthermore, we figured out that uniformly selected training data lead to the best prediction performance. The optimal loss functions and method for how to compose training data studied in this work would be applied to engineering problems such as evaluating DNA length, analyzing biomolecules, predicting concentration of colloidal suspension.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.46 no.4
• /
• pp.392-405
• /
• 2010

The population of small yellow croaker, Larimichthys polyactis, in the Yellow and East China Seas has decreased significantly since the mid 1970s. Several management measures have been introduced to conserve it, but population size remains low. To rebuild this population, it is now necessary to consider more effective management methods based on the stock assessment. To determine long-term population changes, fishery and biological data collected over 34 years (1969-2002) were analysed. Yearly fish length compositions were analysed for the time periods 1968 through 1970, 1978 through 1982, and 1993 through 2002; and catch data was available from 1969 to 2002. Annual population sizes were calculated based on length composition, the relationship between total length and body weight, and total landings. Analyses showed that since the 1970s, average size of harvested fish decreased; the proportion of less mature fish (smaller than the 50% maturity length, 19cm) in catches has increased and the estimated biomass has decreased significantly. Consequently, the main management recommendation is that juvenile fish need to be better protected to allow the rebuilding of resources to a more sustainable population level. This will require fish size limit, permissible mesh size, and closed area and season regulations.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.50 no.4
• /
• pp.556-566
• /
• 2014

Korean distant water tuna longline fishery commenced in 1966 in the Atlantic Ocean. Since then, it has become one of the most important fisheries of Korea. By early of 1970s, total amount of tuna caught in the Atlantic Ocean was the highest among other Korean distant water tuna fisheries, but has become minor since 1990s. The annual catch of tuna and tuna-like species by Korean tuna longline fishery in the Atlantic Ocean was about 1,900 mt in 2013. Bigeye tuna was the predominant species in species composition followed by yellowfin tuna, Albacore tuna, Blue shark and Swordfish. Korean distant water tuna longline fishery have mainly operated in the tropical area of the Atlantic Ocean ($20^{\circ}N{\sim}20^{\circ}S$, $20^{\circ}E{\sim}60^{\circ}W$), fishing ground was almost similar as in the previous years. The length frequency of major species (Bigeye tuna, Yellowfin tuna, Albacore tuna, Blue shark and Swordfish) were estimated. As the result of length (size) frequency data on main species caught by lonline fishery in the Atlantic Ocean, main length intervals of bigeye tuna caught in 2011, 2012 and 2013 were 120~125 cm, 160~165 cm and 130~135 cm, respectively. For yellowfin tuna, those were 125~130 cm, 150~155 cm, 145~150 cm and for albacore, 109 cm, 102 cm, 109 cm and 106 cm respectively. For swordfish caught in 2011, 2012 and 2013, main length intervals were 130~135 cm, 125~135 cm and 125~130 cm, respectively, and for blue shark, 195~200 cm in 2011 and 185~190 cm in 2012, 2013.

• The Journal of the Society of Korean Medicine Diagnostics
• /
• v.17 no.2
• /
• pp.112-121
• /
• 2013

Objectives This study analyze the correlation between iris markings and the obesity level shown in body composition indices. And It quantitatively evaluates iris markings according to the level of obesity to improve a clinical application plan for the iris analysis. Methods This study analyzed the iris and body composition through the quantization of iris markings for 28 female patients. It calculated descriptive statistics by using SPSS 13.0 and investigated correlation coefficients of non-parametric Spearman in order to analyze the relation between iris markings and body composition indices. Results Considering the correlation between iris markings and body composition indices, an area of the pupil had a significant positive correlation with height and with fat-free amount such as body water, protein, and minerals. BMI showed significant positive correlations with the length of the autonomic nerve wreath and the ratio of the autonomic nerve wreath, and the area of the autonomic nerve wreath. Moreover, the number of lacunas in the pancrea area had a significant positive correlation with weight and BMI. Conclusions This study objectifies the quantification of iris markings to identify the correlation between iris markings and the level of obesity. The study results will be used as a basic data to make standards for the diagnosis of obesity, the treatment of related diseases, and the diagnosis of iris markings for prevention.

• Ocean Science Journal
• /
• v.42 no.2
• /
• pp.69-79
• /
• 2007

A vertical transect with 4 km length was established for the macrofaunal survey on the Chokchon macrotidal flat in Kyeonggi Bay, Incheon, Korea, 1994. Tidal elevation (m) and sediment mean grain size $(\phi)$ were inversely predicted by the transfer functions from the faunal assemblages. Three methods: weighted average using optimum value (WA), tolerance weighted version of the weighted average (WAT) and maximum likelihood calibration (MLC) were employed. Estimates of tidal elevation and mean grain size obtained by using the three different methods showed positively corresponding trends with the observations. The estimates of MLC were found to have the minimum value of sum of squares due to errors (SSE). When applied to the previous data $(1990\sim1992)$, each of three inference models exhibited high predictive power. This result implied there are visible relationships between species composition and faunas' critical environmental factors. Although a potential significance of the two major abiotic factors was re-affirmed, a weak tendency of biological interaction was detected from faunal distribution patterns across the flat. In comparison to the spatial and temporal patterns of the estimates, it was suggested that sediment characteristics were the primary factors regulating the distribution of macrofaunal assemblages, rather than tidal elevation, and the species composition may be sensitively determined by minute changes in substratum properties on a tidal flat.

• Plant Resources
• /
• v.8 no.3
• /
• pp.244-249
• /
• 2005

The growth charateristics and karyotypes of Aster spathulifolius collected from 5 sites including coastal and island region on the Korean peninsula, were analysed. Several morphological characteristics of the plants such as leaf length, leaf width, top internode, medium internode, spike branching, flower diameter, number of petal, leaf color, leaf form, stem and leaf hair, viscosity, and serration of the plants were distinctly different depending on the native region from which they were collected. Karyotypic analysis showed that the chromosome number was all diploid (2n=18), with one pair of submetacentric satellite chromosomes. The chromosome composition included 7 pairs of metacentric chromosomes and 2 pairs of submetacentric chromosomes in all plants. However, chromosome order and the ranges of the chromosome lengths were a little different from plant to plant according to their native growing regions. The plants from Geoje-Do especially showed large differences in the chromosome lengths between the longest and the shortest compared to the plants from other places. This results provide important data to support the classification of the species into several sub-species.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.55 no.5
• /
• pp.590-597
• /
• 2022

We analyzed the diet composition of smallmouth argentine Argentina kagoshimae, which were collected via bottom trawl in the South Sea during February, April, August, and October 2021. Their size range was 8.1-21.4 cm total length. A. kagoshimae are mesopelagic carnivores, mainly consuming benthopelagic Euphausiacea and Caridea crustaceans. Polychaetes were the second largest prey component in their diet, which also included small quantities of copepods, amphipods, and arrow worms. Stomach content compositions substantially differed among body size levels in A. kagoshimae. The results of non-metric multidimensional scaling ordination and permutational multivariate analysis of variance of dietary data revealed no significant dietary differences. The results of a graphical analysis of the diet composition showed that A. kagoshimae are specialized feeders characterized by strong individual feeding specialization.

• Journal of Korean Society of Water and Wastewater
• /
• v.13 no.3
• /
• pp.61-72
• /
• 1999

To obtain the experimental data for design and operation of actual filtration processes, a sand filter and three kinds of dual media filters in pilot-plant scale were operated in this study. We analyzed the effect of filter medium composition on the filter performance and the effects of backwash water flow rates, length of stream line and air flow rate on the filter backwash efficiency. We also compared the efficiencies of the combined air-water backwashing and the water backwashing in dual media filters. As the backwash water flow rates or the length of stream line increased, the final turbidity of backwash water was decreased and the filtration duration time after backwash was increased. In the case of the combined air-water backwashing, the backwash water quantity needed for backwashing the dual media filters could be decreased. The total volume of filtered water for the dual media filters during filter run was over three times larger than that for the sand filter. The dual media filters could be operated at a high filtration rate of 360 m/day.

• Nuclear Engineering and Technology
• /
• v.52 no.12
• /
• pp.2709-2716
• /
• 2020

Compositions of large nuclear cores (e.g. boiling water reactors) are highly heterogeneous in terms of fuel composition, control rod insertions and flow regimes. For this reason, they usually lack high order of symmetry (e.g. 1/4, 1/8) making it difficult to estimate their neutronic parameters for large spaces of possible loading patterns. A detailed hyperparameter optimization technique (a combination of manual and Gaussian process search) is used to train and optimize deep neural networks for the prediction of three neutronic parameters for the Ringhals-1 BWR unit: power peaking factors (PPF), control rod bank level, and cycle length. Simulation data is generated based on half-symmetry using PARCS core simulator by shuffling a total of 196 assemblies. The results demonstrate a promising performance by the deep networks as acceptable mean absolute error values are found for the global maximum PPF (~0.2) and for the radially and axially averaged PPF (~0.05). The mean difference between targets and predictions for the control rod level is about 5% insertion depth. Lastly, cycle length labels are predicted with 82% accuracy. The results also demonstrate that 10,000 samples are adequate to capture about 80% of the high-dimensional space, with minor improvements found for larger number of samples. The promising findings of this work prove the ability of deep neural networks to resolve high dimensionality issues of large cores in the nuclear area.

• Korean Journal of Ecology and Environment
• /
• v.54 no.1
• /
• pp.12-23
• /
• 2021

As fish communities are sensitive to natural environmental changes as well as anthropogenic pressures, their composition and characteristics help us to assess the aquatic ecosystem health. The fish fauna and the length-weight relationship of Zacco platypus in the Seomjin River system were investigated from July 2018 to May 2019. The collected species during the survey period were 49 species belong to 14 families including 19 Korean endemic and two endangered species. According to Bray-Curtis similarities, 14 sites were divided into four groups based on the fish community composition; two groups(group A, B) and two uncategorized sites(group C, D). There were significant differences between groups A and B(analysis of similarities, R=0.722, p=0.002). A similarity percentage analysis revealed that Z. koreanus(8.55%), Micropterus salmoides(6.90%), and Lepomis macrochirus (4.90%) contributed to these differences of fish assemblages between group A and B. Group A showed higher relative abundances of exotic species such as M. salmoides and L. macrochirus, while Z. platypus and Z. koreanus which are the common species in Korea, were the most dominant species in group B. Based on the length-weight relationship of Z. platypus, b values were varied from 2.82 to 3.80, indicating that growth patterns of the species were spatially different. This study could be served as baseline data for understanding fish fauna, assessing habitat characteristics based on the fauna, and identifying health conditions of Z. platypus in the Seomjin River system.