• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.85-92
• /
• 2020

Deep learning shows outstanding performance in image and video analysis, such as object classification, object detection and semantic segmentation. In this paper, it is analyzed that the performances of deep learning models can be affected by characteristics of train dataset. It is proposed as a method for selecting activation function and optimization algorithm of deep learning to classify facial expression. Classification performances are compared and analyzed by applying various algorithms of each component of deep learning model for CK+, MMI, and KDEF datasets. As results of simulation, it is shown that genetic algorithm can be an effective solution for optimizing components of deep learning model.

• The Journal of Industrial Distribution & Business
• /
• v.10 no.2
• /
• pp.25-37
• /
• 2019

Purpose - The goal of this study is to analyze the differences in education performances between students of the government's financial support program and those who do not receive support at a local university in Korea. Research design, data, and methodology - The questionnaire used was NASEL. NASEL is considered a highly suitable survey tool for professors, courses, and performances in Korean universities. The 290 students who participated and 44 students do not participate in the financial support program were surveyed for 10 days. The characteristics of students were investigated by frequency analysis and technical statistics. The analysis of student collective characteristics used independent t and f-tests,and one-way ANOVA with IBM SPSS Statistics 22.0 for statistical purposes. Results - The p-value of the group receiving financial support and the group without financial support in active-collaborative learning is 0.167. The p-value of the economically supported group and the non-supported group of the faculty-student interaction is 0.281. The confidence coefficient of the active-collaborative learning questionnaire is 0.861. The reliability coefficient of the questionnaire for the faculty-student interaction questionnaire is 0.871. Conclusions - There are no clear differences in active-collaborative learning and faculty-student interaction between participating and non-participating students in the economic program.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.20 no.2
• /
• pp.471-476
• /
• 2006

Chongmyung-tang(CMT, 聰明湯), oriental herbal medicine which consists of Polygaglae Radix(遠志), Acori Graminei Rhizoma(石菖蒲) and Hoelen(白茯神) has effect on amnesia, dementia. In order to evaluate effect of CMT on memory and learning in mice, CMT extract was used for studies. This paper describes the effects of CMT extract on memory and learning processes by using the passive and active avoidance performance tests, novel object recognition task and water maze task. The CMT extract ameliorated the memory retrieval deficit induced by ethanol in the passive avoidance responses but did not affect ambulatory activity of normal mice. These results suggest that CMT has an ameliorating effect on memory retrieval impairment. CMT extract decreased spontaneous motor activity(SMA) in the latter sessions of memory registration in active avoidance responses. These results suggest that CMT has partly transquilizing or antianxiety effects. In novel object recognition task to measure visual recognition memory, CMT-administered mice enhanced in long term memory for 1-3 days. In water maze task to measure spatial learning, which requires the activation of NMDA receptors in the hippocampus, spatial learning in CMT-administered mice was faster than in wild-type mice. These results suggest that CMT enhances memory and activates NMDA receptors.

• Korean Journal of Cognitive Science
• /
• v.13 no.3
• /
• pp.47-60
• /
• 2002

The study investigated the effects of types of presentation and cognitive load on multimedia learning. In experiment 1, subject were 90 elementary school students. The subject were assigned in three conditions: Narration and Text (NT) condition, Animation and Narration(AN) condition, Animation and Text(AT) condition. The result showed that AN condition improved the learning performances in comparison with AT condition, NT condition. Experiment 2 was administrated to 87 undergraduate students. They were participated in three conditions, also. The conditions were Animation and Text (AT) condition, Animation and Narration (AN) condition, Animation, Narration and Text (ANT) condition. the results showed that AN condition was greater in AT, ANT condition. The results from a series of these experiments imply that varying the types of presentation of identical learning materials had influences on the performances. Multimedia presentation(animation and verbal conditions) improved the learning performances in comparison with monomedia presentation(verbal condition), and the advantage was raised when learners were provided the learning material in the multimodal and multimedia environment(AN condition). Also, it came out that redundant text identical to narration disrupted learning when learners were in the picture (either animation or illustration) and narration conditions. Likewise, also for adults, performances were improved in the multimodal conditions and redundant text identical to narration was not helpful for learning. These results are evidences for the dual-coding theory and the cognitive load theory.

• The Journal of Korean Medicine
• /
• v.25 no.4
• /
• pp.51-60
• /
• 2004

Objective : The effect Saenggitang (GT), which has been used for amnesia, in Oriental Medicine, on memory and learning ability, was investigated. Methods : Hot water extracts (HWE) of SGT were used for the studies. In passive avoidance performances (step through test), active avoidance performances (lever press test), Motor activity, pentobarbital-induced sleep, 20 and 50 mg/100g of SGT-HWE ameliorated the memory retrieval deficit induced by 40% ethanol. Results : The SGT-HWE did not affect the ambulatory activity of normal mice in normal condition. 20 and 50 mg/100g of SGT-HWE enhanced contextual fear memory, but not cued fear memory in a fear conditioning task, which requires the activation of the NMDA (N-methyl-D-aspartase) receptor. SGT-HWE did not affect the motor activity measured by the titling type ambulometer test performed immediately and 24 hr after the administration. SGT-HWE prolonged the sleeping time induced by 50 mg/kg pentobarbital in mice and decreased SMA (spontaneous motor activity) in active avoidance performances (lever press test). Conclusion : These results indicate that the SGT-HWE have an improving effect on the memory retrieval disability induced by ethanol and may act as a stimulating factor for activating the NMDA receptor. and the SGT-HWE has a tranquilizing and anti-anxiety action.

• Korean Journal of Construction Engineering and Management
• /
• v.20 no.5
• /
• pp.31-41
• /
• 2019

The purpose of this study is to investigate the participation of vocational training program of construction engineers and the impact of workplace learning (formal learning and informal learning) on human resource performances of construction engineers. The data of 306 construction engineers were extracted from 10,069 workers in various industries those were collected by 6th human resource company panel survey done by Korea Research Institute of Vocational Education & Training. This study found that, compared with workers in other industries, participation rate of construction engineers in workplace learning (formal learning, informal learning) was relatively low, and especially the participation rate of informal learning was significantly low. Regression analysis showed that participation in formal learning did not affect positive job performance and job satisfaction. On the other hand, informal learning has a positive effect on job capability, job satisfaction, and organizational commitment.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.11
• /
• pp.3099-3120
• /
• 2023

The present fast-moving era brings a serious stress issue that affects elders and youngsters. Everyone has undergone stress factors at least once in their lifetime. Stress is more among youngsters as they are new to the working environment. whereas the stress factors for elders affect the individual and overall performance in an organization. Electroencephalogram (EEG) based stress level classification is one of the widely used methodologies for stress detection. However, the signal processing methods evolved so far have limitations as most of the stress classification models compute the stress level in a predefined environment to detect individual stress factors. Specifically, machine learning based stress classification models requires additional algorithm for feature extraction which increases the computation cost. Also due to the limited feature learning characteristics of machine learning algorithms, the classification performance reduces and inaccurate sometimes. It is evident from numerous research works that deep learning models outperforms machine learning techniques. Thus, to classify all the emotions based on stress level in this research work a hybrid deep learning algorithm is presented. Compared to conventional deep learning models, hybrid models outperforms in feature handing. Better feature extraction and selection can be made through deep learning models. Adding machine learning classifiers in deep learning architecture will enhance the classification performances. Thus, a hybrid convolutional neural network model was presented which extracts the features using CNN and classifies them through machine learning support vector machine. Simulation analysis of benchmark datasets demonstrates the proposed model performances. Finally, existing methods are comparatively analyzed to demonstrate the better performance of the proposed model as a result of the proposed hybrid combination.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.1
• /
• pp.90-98
• /
• 2015

In this paper, an advanced teaching-learning based optimization(TLBO) method for the magnetic levitation controller of Maglev transportation system is proposed to optimize the control performances. An attraction-type levitation system is intrinsically unstable and requires a delicate control. It is difficult to completely satisfy the desired performance through the methods using conventional methods and intelligent optimizations. In the paper, we use TLBO and clonal selection algorithm to choose the optimal control parameters for the magnetic levitation controller. To verify the proposed algorithm, we compare control performances of the proposed method with the genetic algorithm and the particle swarm optimization. The simulation results show that the proposed method is more effective than conventional methods.

• Journal of Distribution Research
• /
• v.17 no.3
• /
• pp.93-113
• /
• 2012

Early conflict research in channel and organization area have focused on the definition of conflict construct, its cause, consequence and identified conflict resolution management. Recent studies about conflict, however, have explored new assumption of complexity, a multidimensional conflict construct, contextual conflict management strategies, positive and negative conflict/consequence, and the conflict resolution strategy. Although many literatures exists on channel conflict resolution, little research has been done about relationship learning and performance from conflict resolution perspective. This study explores how channel members can achieve a relationship learning, as a conflict resolution mechanism, which enhance co-created value in marketing channel relationship. Therefore we propose that conflict resolution strategies(collaborating behavior and avoiding behavior) influence channel performance(effectiveness and efficiency) through relationship learning processes(learning via information exchange, joint interpretation and coordination, relationship-specific knowledge memory), in view of buyer-seller relationship. The research model is shown at

• Communications for Statistical Applications and Methods
• /
• v.30 no.2
• /
• pp.215-226
• /
• 2023

Deep learning has made great strides in the field of unstructured data such as text, images, and audio. However, in the case of tabular data analysis, machine learning algorithms such as ensemble methods are still better than deep learning. To keep up with the performance of machine learning algorithms with good predictive power, several deep learning methods for tabular data have been proposed recently. In this paper, we review the latest deep learning models for tabular data and compare the performances of these models using several datasets. In addition, we also compare the latest boosting methods to these deep learning methods and suggest the guidelines to the users, who analyze tabular datasets. In regression, machine learning methods are better than deep learning methods. But for the classification problems, deep learning methods perform better than the machine learning methods in some cases.