• Ecology and Resilient Infrastructure
• /
• v.10 no.4
• /
• pp.107-115
• /
• 2023

High turbidity in source water can have adverse effects on water treatment plant operations and aquatic ecosystems, necessitating turbidity management. Consequently, research aimed at predicting river turbidity continues. This study developed a multi-class classification model for prediction of turbidity using LightGBM (Light Gradient Boosting Machine), a representative ensemble machine learning algorithm. The model utilized data that was classified into four classes ranging from 1 to 4 based on turbidity, from low to high. The number of input data points used for analysis varied among classes, with 945, 763, 95, and 25 data points for classes 1 to 4, respectively. The developed model exhibited precisions of 0.85, 0.71, 0.26, and 0.30, as well as recalls of 0.82, 0.76, 0.19, and 0.60 for classes 1 to 4, respectively. The model tended to perform less effectively in the minority classes due to the limited data available for these classes. To address data imbalance, the SMOTE (Synthetic Minority Over-sampling Technique) algorithm was applied, resulting in improved model performance. For classes 1 to 4, the Precision and Recall of the improved model were 0.88, 0.71, 0.26, 0.25 and 0.79, 0.76, 0.38, 0.60, respectively. This demonstrated that alleviating data imbalance led to a significant enhancement in Recall of the model. Furthermore, to analyze the impact of differences in input data composition addressing the input data imbalance, input data was constructed with various ratios for each class, and the model performances were compared. The results indicate that an appropriate composition ratio for model input data improves the performance of the machine learning model.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.21 no.8
• /
• pp.857-864
• /
• 2010

In this paper, we analyse the effect of the path imbalances(gain and phase mismatches) in LINC(LInear amplification with Nonlinear Component) system, and propose a simple scheme using LUTs(Look Up Table) to compensate the path imbalances. The EVM(Error Vector Magnitude) and ACPR(Adjacent Channel Power Ratio) of the LINC system are degraded significantly by the path imbalances because it adopts an outphasing technique. The EVM and ACPR are theoretically extracted for two variables(gain and phase mismatch factors) and 2-D LUTs for those are generated based on the analysis. The efficient and simple compensation scheme for the path imbalances is proposed using the 2-D LUTs. A LINC system with the suggested compensation scheme is implemented, and the proposed method is verified with an experiment. A 16-QAM signal with 1.5 MHz bandwidth is used. Before the compensation, the path gain ratio was 95 % and phase error was $19.33^{\circ}$. The proposed scheme adjusts those values with 99 % and $0.5^{\circ}$, and improves ACPR about 18.1 dB.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Journal of Environmental Science International
• /
• v.24 no.12
• /
• pp.1649-1656
• /
• 2015

Jeju Island is the highest rain-prone area in Korea that possesses affluent water resources, but future climate changes are predicted to further increase vulnerabilities as resultant of increasing of extreme events and creating spatial-temporal imbalance in water resources. Therefore, this study aimed to provide basic information to establish a proper water resources management plan by evaluating the effects of climate change on water resources using climate change scenario. Direct runoff ratio for 15 years (2000~2014) was analyzed to be 11~32% (average of 23%), and average direct runoff ratio for the next 86 years (2015~2100) was found as 28%, showing an increase of about 22% compared to the present average direct runoff ratio (23%). To assess the effects of climate change on long-term runoff, monthly runoff variation of future Gangjeong watershed was analyzed by dividing three time periods as follows: Present (2000~2030), Future 1 (2031~2070) and Future 2 (2071~2100). The estimated results showed that average monthly runoff increases in the future and the highest runoff is shown by Future 2. Extreme values has been expected to occur more frequently in the future as compared to the present.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.7
• /
• pp.3393-3412
• /
• 2017

As a main challenge in network virtualization, virtual network embedding problem is increasingly important and heuristic algorithms are of great interest. Aiming at the problems of poor correlation in node embedding and link embedding, long distance between adjacent virtual nodes and imbalance resource consumption of network components during embedding, we herein propose a two-stage virtual network embedding algorithm NA-PVNM. In node embedding stage, resource requirement and breadth first search algorithm are introduced to sort virtual nodes, and a node fitness function is developed to find the best substrate node. In link embedding stage, a path fitness function is developed to find the best path in which available bandwidth, CPU and path length are considered. Simulation results showed that the proposed algorithm could shorten link embedding distance, increase the acceptance ratio and revenue to cost ratio compared to previously reported algorithms. We also analyzed the impact of position constraint and substrate network attribute on algorithm performance, as well as the utilization of the substrate network resources during embedding via simulation. The results showed that, under the constraint of substrate resource distribution and virtual network requests, the critical factor of improving success ratio is to reduce resource consumption during embedding.

• Journal of Korean Society of Rural Planning
• /
• v.25 no.4
• /
• pp.99-107
• /
• 2019

This paper analyzed the characteristics of population change from 2000 to 2018 in 466 mountainous areas using resident registration data from the Ministry of the Interior and Safety, and projected the population in those areas through 2050 with the cohort change ratio method. The population had dramatically decreased from 2000 to 2009. With the slowing population decrease after 2010, the population has increased gradually since 2014. Especially the population of ages over 65 in 2018 had increased 34% compared to 2000, while the working age population had decreased 29%. This shows that population aging becomes serious problems in the mountainous area. Assuming the cohort change ratios from 2010 to 2015 and child-woman ratio in 2015 remain constant, it appeared that the projected population of the mountainous area dropped to 1.26 million in 2030 and 820,000 in 2050. It is expected to have a population with an inverted pyramid structure showing a gender imbalance with more females in 60's and 70's. Although it continues to show the recent population growth in mountainous area, population in mountainous area is expected to consistently decrease. Therefore, it is required to develop policies and strategies to promote an influx of people into mountainous area for maintaining functionality and sustainability of mountainous areas.

• The Journal of Korean Academy of Orthopedic Manual Physical Therapy
• /
• v.11 no.2
• /
• pp.71-82
• /
• 2005

The couple force of the scapular stabilizers for upward rotation that include the upper and lower trapezius, serratus anterior muscles is essential to maintain the stability of shoulder joint. But if there is an imbalance of these muscles, it would occur many problems of shoulder joint. A push-up plus exercise with an unstable support is used in order to increase the muscle activity and stimulate the proprioception of shoulder joint. The purpose of this study was to compare the muscle activity of upper and lower trapezius, serratus anterior muscles and to determine which exercise is more effective when subjects perform push-up plus exercise in the stable support and unstable support. 15 healthy subjects and 15 painful subjects with injury of shoulder joint participated in this study. Surface electromyography data were collected during plus phase of push-up plus exercise. The types of push-up plus exercise were composed of three conditions. Stable type is subjects performed the push-up plus exercise on the fixed support and unstable type 1 is on the sling without shaking and unstable type 2 is on the sling with shaking by tester's manual. The upper and lower trapezius activities of injured group were higher than uninjured group at three measure conditions, but serratus anterior activities were not. The UT/SA ratio of injured group was higher than uninjured group at three measure conditions. The serratus anterior activities were most high at unstable 2 measure in both groups. The UT/SA ratio was most low at unstable 2 measure condition in injured group only. The present result revealed that push-up plus exercise in the unstable support with shaking which took most high serratus anterior activities and most low of UT/SA ratio is optimal cure method that can improve the imbalance of the scapular stabilizer.

• Journal of Korean Ophthalmic Optics Society
• /
• v.15 no.4
• /
• pp.398-404
• /
• 2010

Purpose: Concerns had been voiced about an oversupply of optometrists in Korea. So, this study aimed to forecast the supply and demand for opticians for policy implications. Methods: Baseline Projection model combined with demographic method was adopted as the supply forecasting method and so was a ratio method using the number of physician and population using weight of healthcare utilization. Results: Under the 'physician to optician ratio', there would be a surplus of 83~700 opticians in 2010 and an undersupply of 15 to surplus of 6,118 opticians in 2025. Under the 'population to optician ratio', there would be a surplus of 1,055 opticians in 2010 and surplus of 9,376 opticians in 2025. Conclusions: We concluded that there would be oversupply for opticians until 2025, although the shortage and surplus of opticians might depend on the ratio's criteria. Hence, policies would need to be developed that could solve the imbalance in requirements and supply for opticians.

• Journal of International Academy of Physical Therapy Research
• /
• v.11 no.2
• /
• pp.2071-2076
• /
• 2020

Background: Based on the understanding of the muscle activation relationship between the infraspinatus and posterior deltoid muscles to according to the angle of motion during external rotation on glenohumeral joint, effective shoulder joint strengthening exercise for the prevention and rehabilitation of shoulder injury due to muscle strength imbalance can be performed by achieving the ideal muscle activity ratio during exercise. Objectives: To compare and analyze the muscle activation changes and activity ratio of the infraspinatus and posterior deltoid muscles according to the glenohumeral external rotation angle. Design: Quasi-randomized trial. Methods: The study included 48 healthy male and female adults who provided informed consent for participation in the study. All the subjects performed isometric glenohumeral external rotation by setting the angle of motion to 30°, 45°, and 60° using a 5 kg resistance weight pulley. On surface electromyography, the differences in muscle activation and activity ratio between the infraspinatus and posterior deltoid muscles were investigated. Results: A significant difference in muscle activation was found in the comparison between the infraspinatus and posterior deltoid muscles according to the glenohumeral external rotation angle (P<.05). The muscle activation levels of the infraspinatus and posterior deltoid muscles were highest at the external rotation angles of 30° and 60°, respectively. The muscle activity ratio between the infraspinatus and posterior deltoid muscles also showed a significant difference (P<.05) and was highest at the shoulder external rotation angle of 30°. Conclusion: The findings of this study suggest that muscle activity is the highest at the shoulder external rotation angle of 30° in healthy individuals.

• The Journal of Korean Academy of Orthopedic Manual Physical Therapy
• /
• v.16 no.1
• /
• pp.26-36
• /
• 2010

Purpose: The study compared the muscle activity and ratio of upper trapezius (UT) to lower trapezius (LT) activity between acute and chronic whiplash-associated disorder (WAD) patients. Methods: Twelve healthy (male: 7), 14 acute WAD (male: 7), and 11 chronic WAD (male: 3) volunteers participated in this study. Electromyography using a surface EMG recorded the activity of the upper trapezius and lower trapezius of both shoulders (dominant and non-dominant) during $120^{\circ}$ elevation when standing and shoulder depression when sitting. The testing order was selected randomly. Subjects were asked to maintain each experimental position for 5 seconds at end range. EMG activity was normalized using the maximal voluntary isometric contraction (MVIC) elicited using a manual muscle-testing technique. One-way repeated measures analysis of variance (ANOVA) was used to compare the average root mean square (RMS) value of EMG activity for each condition. Results: The EMG activity of the dominant UT for chronic WAD subjects was significantly higher than for acute WAD subjects during $120^{\circ}$ elevation (p<.05). The EMG activity of the dominant LT for acute WAD subjects was significantly lower than for the control group (p<.05) during $120^{\circ}$ elevation. The EMG activity of the dominant LT for WAD patients was significantly lower than for the control group during shoulder depression (p<.05), and the chronic WAD patients scored significantly the lowest (p<.01). The dominant UT/LT for chronic WAD patients was significantly higher than the ratio for acute WAD subjects during shoulder depression (p<.05); this result was higher than for the control group (p<.01). The non-dominant UT/LT (ratio) for chronic WAD patients was significantly higher than the ratio for acute WAD subjects during shoulder depression (p<.05). Conclusion: The UT for chronic WAD subjects was hyperactive when compared to the acute WAD subjects, and the was hypoactive for both acute and chronic patients, therefore intra-trapezius imbalance was more prevalent during shoulder depression.