• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.818-833
• /
• 2021

To address the problems of low precision rate, insufficient feature extraction, and poor contextual ability in existing text sentiment analysis methods, a mixed model account of a CNN-BiLSTM-TE (convolutional neural network, bidirectional long short-term memory, and topic extraction) model was proposed. First, Chinese text data was converted into vectors through the method of transfer learning by Word2Vec. Second, local features were extracted by the CNN model. Then, contextual information was extracted by the BiLSTM neural network and the emotional tendency was obtained using softmax. Finally, topics were extracted by the term frequency-inverse document frequency and K-means. Compared with the CNN, BiLSTM, and gate recurrent unit (GRU) models, the CNN-BiLSTM-TE model's F1-score was higher than other models by 0.0147, 0.006, and 0.0052, respectively. Then compared with CNN-LSTM, LSTM-CNN, and BiLSTM-CNN models, the F1-score was higher by 0.0071, 0.0038, and 0.0049, respectively. Experimental results showed that the CNN-BiLSTM-TE model can effectively improve various indicators in application. Lastly, performed scalability verification through a takeaway dataset, which has great value in practical applications.

• Wind and Structures
• /
• v.38 no.2
• /
• pp.147-160
• /
• 2024

The aerodynamic force is a significant component that influences the stability and safety of structures. It has unstable properties and depends on computer precision, making its long-term prediction challenging. Accurately estimating the aerodynamic traits of structures is critical for structural design and vibration control. This paper establishes an unsteady aerodynamic time series prediction model using Long Short-Term Memory (LSTM) network. The unsteady aerodynamic force under varied Reynolds number and angles of attack is predicted by the LSTM model. The input of the model is the aerodynamic coefficients of the 1 to n sample points and output is the aerodynamic coefficients of the n+1 sample point. The model is predicted by interpolation and extrapolation utilizing Unsteady Reynolds-average Navier-Stokes (URANS) simulation data of flow around a circular cylinder, square cylinder and airfoil. The results illustrate that the trajectories of the LSTM prediction results and URANS outcomes are largely consistent with time. The mean relative error between the forecast results and the original results is less than 6%. Therefore, our technique has a prospective application in unsteady aerodynamic force prediction of structures and can give technical assistance for engineering applications.

• The Journal of the Korea Contents Association
• /
• v.13 no.1
• /
• pp.342-353
• /
• 2013

The purpose of this study was to investigate the effects of short-term Maum Meditation and to compare effects according to the length of the training period on mental health of teachers. In this research, 2 types of design were used which were one-group pre-posttest design and descriptive study. In short-term Maum Meditation group, 2 times of surveys were carried out, 75 teachers in the faculty training filled out the structured questionnaire. In Long-term Maum Meditation group, 101 teachers who were attending Maum Meditation above one-year filled out the structured questionnaire. The data were analyzed by SPSS WIN 14.0 program. The results showed that there was a significant difference in mental health including 9 subscales for short-term Maum Meditation group. Also, in long-term Maum Meditation group, there were significant differences in mental health excepting paranoid ideation. It was founded that Maum Meditation is the way for the mental health promotion in short-term and long-term period. So it is necessary to develop program according to training period for not only clinical application but also to strengthen coping ability for stress.

• The Journal of the Korea Contents Association
• /
• v.11 no.12
• /
• pp.943-949
• /
• 2011

Smart scope, such as smart equipments, smart homes, and smart phones, has been appeared lately being developed. Such technology developments have been close and practical to our daily lives, not regulated in a science fiction and a movie. It is just the peak of a smart term period. However, with the result from imprudent application of the term 'smart' in the various fields in a short term, the concept of the term 'Smart City' has been nebulous. Therefore, this study is intended to design the obvious concept of Smart City and figure out the political improvement how it will make the direction in the future. That is to study the Smart City which applies the term smart in its city and space, and to examine the improvement directions. After studying the application of the term 'smart', we determine this word has intelligent and high-tech concepts and is used variously in the diverse fields. But we conclude that the term Smart City in these fields is defined differently. From these perspectives, the term is not a simple creature consists of 'smart' and 'city.' It is to be designed as the meanings of making the cities smart.

• Proceedings of the Korean Geotechical Society Conference
• /
• 1999.10a
• /
• pp.481-488
• /
• 1999

Recently a removable soil nail is demanded due to problems beyond of economical and engineering purpose. In this study controlled displacement and controlled force field pull-out tests are carried out 7 times in order to evaluate short-term and long-term pull-out characteristics of the removable soil nail. For evaluating application of removable soil nailing system, bending tests of removable soil nails and tensile tests of fixed sockets are carried out. In the removable soil nailing system, the predicted horizontal displacements from FLAC-2D are also compared with the field measurements occurred in stepwise excavation. And approach for the stability analysis of removable soil nailing system after removed is proposed.

• Smart Structures and Systems
• /
• v.30 no.6
• /
• pp.649-659
• /
• 2022

Structural Health Monitoring (SHM) researchers have identified Augmented Reality (AR) as a new technology that can assist inspections. Post-seismic structural inspections are conducted to evaluate the safety level of the damaged structures. Quantification of nearby structural changes over short-term and long-term periods can provide building inspectors with information to improve their safety. This paper proposes a Time Machine Measure (TMM) application based on an Augmented Reality (AR) Head-Mounted-Device (HMD) platform. The primary function of TMM is to restore the saved meshes of a past environment and overlay them onto the real environment so that inspectors can intuitively measure dynamic structural deformation and other environmental movements. The proposed TMM application was verified by demo experiments simulating a real inspection environment.

• Journal of The Korea Institute of Healthcare Architecture
• /
• v.19 no.1
• /
• pp.9-22
• /
• 2013

Purpose: The aging population is the quickly increasing in Korea. Since 2008, the National Long term Care Insurance, a number of long term care facilities have established during short time in Korea. Especially, the environmental assessment tool is important for managing healing environment in a long term care facility for the elderly. Alzheimer's Garden Audit Tool (AGAT) is used to assess whether a garden incorporates those elements and qualities as healing outdoor space. In this paper, we discuss the benefits, limitations, and future directions of the assessment tool for long-term care facilities. Methods : The AGAT audit was done through content analysis by 5 experts from diverse discipline as post occupancy evaluation of a long term care facility located in suburban area. The expert group was asked to describe their comments for modification and improvement in application of AGAT. Results : The results of this study show that it is necessary to refine the instrument's items to better meet the needs of the criteria and items for Korean culture-friendly tool. Especially, greenhouse elements (various plants, birds etc.), multisensory experiences, users-oriented space, programmed activities in garden, and hazards on pathway has somewhat limitation as measurement to assess outdoor of long-term care facilities in Korea. Implications : Consequently, AGAT could be applicable to evaluate the outdoor space of long term care facility in Korea with culturally sensitive revision.

• Journal of IKEEE
• /
• v.18 no.4
• /
• pp.442-446
• /
• 2014

A method for applying emergency ratings to improve the reliability of power supply in ACSR overhead transmission lines is described in this paper. Due to re-regulate power industry, most power companies worldwide as well KEPCO have been searching for only economical strategies without new investment. Power demand was rapidly increasing, however, generation amount did not follow sufficiently. Hence in order to increase the transmission capacity for the existing transmission lines in case of peak load, or contingency in transmission lines, an application method of emergency ratings such as short or long term rating is proposed. If applying long term emergency rating instead of static line rating for the period of a peak load, power transmission can be increased to 10 % or more. Furthermore, it was shown that emergency rating can be effectively used in the contingency of double-circuit transmission lines and/or overload cases.

• Progress in Superconductivity and Cryogenics
• /
• v.6 no.2
• /
• pp.25-28
• /
• 2004

Three-phase inductive superconducting fault current limiter (SFCL) with DC reactor rated on 6.6 $KV_{rms}/200 A_{rms}$ has been developed in Korea. This system consists of one DC reactor, AC/DC power converter, and a three-phase transformer, which is called magnetic core reactor (MCR). This paper deals with the short-circuit analysis of the SFCL. The DC reactor was the HTS solenoid coil whose inductance was 84mH. The power converter was performed as the dual-mode operation for dividing voltage between the rectifying devices. The short-term normal operation (1 see) and short-circuit tests (2∼3 cycles) of this SFCL were performed successfully. In regular short-circuit test, the fault current was limited as 30% of rated short-circuit current at 2 cycles after the fault. The experimental results have a very similar tendency to the simulation results. Using the technique for the fault detection and SCR firing control, the fault current limiting rate of the SFCL was improved. From this research, the parameters for design and manufacture of large-scale SFCL were obtained.

• Journal of Institute of Control, Robotics and Systems
• /
• v.14 no.12
• /
• pp.1232-1237
• /
• 2008

This paper introduces an application of wavelet analysis to the sensor fusion of GPS/INS/baroaltimeter. Using wavelet analysis the baro-inertial altitude is decomposed into the low frequency content and the high frequency content. The high frequency components, 'details', represent the perturbed altitude change from the long time trend. GPS altitude is also broken down by a wavelet decomposition. The low frequency components, 'approximations', of the decomposed signal address the long-term trend of altitude. It is proposed that the final altitude be determined as the sum of both the details of the baro-inertial altitude and the approximations of GPS altitude. Then the final altitude exclude long-term baro-inertial errors and short-term GPS errors. Finally, it is shown from the test results that the proposed method produces continuous and sensitive altitude successfully.