• Korean Journal of Health Education and Promotion
• /
• v.22 no.2
• /
• pp.19-32
• /
• 2005

Objectives: The purpose of this study was to examine teachers' safety education activities to determine the significant educational diagnosis variables and to identify their needs of safety education in early-child care centers based on the PRECEDE model. Methods: A total of 304 teachers in early-child care centers participated in this study selected by a multi-stage stratified sampling method considering 11 regions in Seoul, Korea. Self-report type questionnaires were posted to all teachers in 220 early-child care centers by ground mailing service and the 304 teachers completed the questionnaires. The participants' responses were anonymously coded into and analyzed in SPSS program. Results: 'Scratch or bite' was the most frequent accident type(78.3%) and the frequent accident places were 'classroom(88.8%)' and 'playground(67.8%)'. The most frequently conducted safety education activities were 'reminding children their safe behaviors at the beginning and the end of daily class' and the next was 'saving a special time for safety education.' For educational diagnosis factors, related to safety education activities, teachers' safety education activity was more frequent when teachers' safety knowledge was high(p<.001), when teachers had good application skills of their knowledge to their teaching activities(p<.001), when they had strong needs on safety training opportunities(p<.05), and their interests on safety education(p<.001). For enabling factors, class preparation by safety education guide-book review(p<.001), by development of educational materials(p<.001), and by search for the related reference (p<.001), and by participation to safety education training programs for teachers(p<.01) were the significant enabling factors on teachers' safety class activities. For the reinforcing factors, the center-wide support of safety education brochures to children (p<.001), the concerns of centers utilizing safety education specialists(p<.001), and the concerns about safety information collection out of centers(p<.001) were significant factors related with teachers' safety education activities. Conclusions: The significant educational and institutional factors on teachers' safety education activities were teachers' concerns on safety education, their interests on safety knowledge, and the strong concerns on child safety education from the centers.

• The KIPS Transactions:PartB
• /
• v.15B no.1
• /
• pp.25-36
• /
• 2008

In this paper, we present a robust hand recognition approach to sudden illumination changes. The proposed approach constructs a background model with respect to hue and hue gradient in HSI color space and extracts a foreground hand region from an input image using the background subtraction method. Eighteen features are defined for a hand pose and multi-class SVM(Support Vector Machine) approach is applied to learn and classify hand poses based on eighteen features. The proposed approach robustly extracts the contour of a hand with variations in illumination by applying the hue gradient into the background subtraction. A hand pose is defined by two Eigen values which are normalized by the size of OBB(Object-Oriented Bounding Box), and sixteen feature values which represent the number of hand contour points included in each subrange of OBB. We compared the RGB-based background subtraction, hue-based background subtraction and the proposed approach with sudden illumination changes and proved the robustness of the proposed approach. In the experiment, we built a hand pose training model from 2,700 sample hand images of six subjects which represent nine numerical numbers from one to nine. Our implementation result shows 92.6% of successful recognition rate for 1,620 hand images with various lighting condition using the training model.

• Korean Chemical Engineering Research
• /
• v.61 no.1
• /
• pp.123-141
• /
• 2023

A sufficient amount of data with quality is needed for training artificial neural networks (ANNs). However, developing ANN models with a small amount of data often appears in engineering fields. This paper presented an ANN model to improve prediction performance of the ammonia emission amount with 83 data. The ammonia emission rate included eleven inputs and two outputs (maximum ammonia loss, N_max and time to reach half of N_max, K_m). Categorical input variables were transformed into multi-dimensional equal-distance variables, and 13 data were added into 66 training data using a generative adversarial network. Hyperparameters (number of layers, number of neurons, and activation function) of ANN were optimized using Gaussian process. Using 17 test data, the previous ANN model (Lim et al., 2007) showed the mean absolute error (MAE) of K_m and N_max to 0.0668 and 0.1860, respectively. The present ANN outperformed the previous model, reducing MAE by 38% and 56%.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.165-184
• /
• 2023

Recently, Deep learning analysis of unstructured text data using language models, such as Google's BERT and OpenAI's GPT has shown remarkable results in various applications. Most language models are used to learn generalized linguistic information from pre-training data and then update their weights for downstream tasks through a fine-tuning process. However, some concerns have been raised that privacy may be violated in the process of using these language models, i.e., data privacy may be violated when data owner provides large amounts of data to the model owner to perform fine-tuning of the language model. Conversely, when the model owner discloses the entire model to the data owner, the structure and weights of the model are disclosed, which may violate the privacy of the model. The concept of offsite tuning has been recently proposed to perform fine-tuning of language models while protecting privacy in such situations. But the study has a limitation that it does not provide a concrete way to apply the proposed methodology to text classification models. In this study, we propose a concrete method to apply offsite tuning with an additional classifier to protect the privacy of the model and data when performing multi-classification fine-tuning on Korean documents. To evaluate the performance of the proposed methodology, we conducted experiments on about 200,000 Korean documents from five major fields, ICT, electrical, electronic, mechanical, and medical, provided by AIHub, and found that the proposed plug-in model outperforms the zero-shot model and the offsite model in terms of classification accuracy.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.5
• /
• pp.102-109
• /
• 2008

In this paper, the robust hybrid modulation type classifier which use both HOS and WT key features and can recognize 10 digitally modulated signals without a priori information in multipath fading channel conditions is proposed. The proposed classifier developed using data taken field measurements in various propagation model (i,e., rural area, small town and urban area) for real world scenarios. The 9 channel data are used for supervised training and the 6 channel data are used for testing among total 15 channel data(i.e., holdout-like method). The Proposed classifier is based on HOS key features because they are relatively robust to signal distortion in AWGN and multipath environments, and combined WT key features for classifying MQAM(M=16, 64, 256) signals which are difficult to classify without equalization scheme such as AMA(Alphabet Matched Algorithm) or MMA(Multi-modulus Algorithm. To investigate the performance of proposed classifier, these selected key features are applied in SVM(Support Vector Machine) which is known to having good capability of classifying because of mapping input space to hyperspace for margin maximization. The Pcc(Probability of correct classification) of the proposed classifier shows higher than those of classifiers using only HOS or WT key features in both training channels and testing channels. Especially, the Pccs of MQAM 3re almost perfect in various SNR levels.

• The KIPS Transactions:PartB
• /
• v.17B no.4
• /
• pp.295-302
• /
• 2010

Example-based super resolution(EBSR) is a method to reconstruct high-resolution images by learning patch-wise correspondence between high-resolution and low-resolution images. It can reconstruct a high-resolution from just a single low-resolution image. However, when it is applied to a text image whose font type and size are different from those of training images, it often produces lots of noise. The primary reason is that, in the patch matching step of the reconstruction process, input patches can be inappropriately matched to the high-resolution patches in the patch dictionary. In this paper, we propose a new patch matching method to overcome this problem. Using an image observation model, it preserves the correlation between the input and the output images. Therefore, it effectively suppresses spurious noise caused by inappropriately matched patches. This does not only improve the quality of the output image but also allows the system to use a huge dictionary containing a variety of font types and sizes, which significantly improves the adaptability to variation in font type and size. In experiments, the proposed method outperformed conventional methods in reconstruction of multi-font and multi-size images. Moreover, it improved recognition performance from 88.58% to 93.54%, which confirms the practical effect of the proposed method on recognition performance.

• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.1-9
• /
• 2022

In the global era, the importance of imported food safety management is increasing. Address information of overseas food companies is key information for imported food safety management, and must be verified for prompt response and follow-up management in the event of a food risk. However, because each country's address system is different, one verification system cannot verify the addresses of all countries. Also, the purpose of address verification may be different depending on the field used. In this paper, we deal with the problem of classifying a given overseas food business address into the administrative district level of the country. This is because, in the event of harm to imported food, it is necessary to find the administrative district level from the address of the relevant company, and based on this trace the food distribution route or take measures to ban imports. However, in some countries the administrative district level name is omitted from the address, and the same place name is used repeatedly in several administrative district levels, so it is not easy to accurately classify the administrative district level from the address. In this study we propose a deep learning-based administrative district level classification model suitable for this case, and verify the actual address data of overseas food companies. Specifically, a method of training using a label powerset in a multi-label classification model is used. To verify the proposed method, the accuracy was verified for the addresses of overseas manufacturing companies in Ecuador and Vietnam registered with the Ministry of Food and Drug Safety, and the accuracy was improved by 28.1% and 13%, respectively, compared to the existing classification model.

• Journal of Korea Water Resources Association
• /
• v.51 no.12
• /
• pp.1217-1227
• /
• 2018

This paper aims to evaluate the applicability of dam inflow prediction model using recurrent neural network theory. To achieve this goal, the Artificial Neural Network (ANN) model and the Elman Recurrent Neural Network(RNN) model were applied to hydro-meteorological data sets for the Soyanggang dam and the Chungju dam basin during dam operation period. For the model training, inflow, rainfall, temperature, sunshine duration, wind speed were used as input data and daily inflow of dam for 10 days were used for output data. The verification was carried out through dam inflow prediction between July, 2016 and June, 2018. The results showed that there was no significant difference in prediction performance between ANN model and the Elman RNN model in the Soyanggang dam basin but the prediction results of the Elman RNN model are comparatively superior to those of the ANN model in the Chungju dam basin. Consequently, the Elman RNN prediction performance is expected to be similar to or better than the ANN model. The prediction performance of Elman RNN was notable during the low dam inflow period. The performance of the multiple hidden layer structure of Elman RNN looks more effective in prediction than that of a single hidden layer structure.

• Journal of the Korea Computer Industry Society
• /
• v.5 no.4
• /
• pp.491-502
• /
• 2004

In this paper, we propose a HMM model using FSVQ(First Section VQ), fuzzy theory and doubly spectral feature, as study on the isolated word recognition system of speaker-independent. In the proposed paper, LPC cepstrum coefficients and regression coefficients of LPC cepstrum as doubly spectral feature be used. And, training data are divided several section and first section is generated codebook of VQ, and then is obtained multi-observation sequences by order of large propabilistic values based on fuzzy nile from the codebook of the first section. Thereafter, this observation sequences of first section is trained and is recognized a word to be obtained highest probaility by same concept. Besides the speech recognition experiments of proposed method, we experiment the other methods under the equivalent environment of data and conditions. In the whole experiment, it is proved that the proposed method is superior to the others in recognition rate.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.1
• /
• pp.68-76
• /
• 2018

Target prioritization is necessary for a multifunction radar(MFR) to track an important target and manage the resources of the radar platform efficiently. In this paper, we consider an artificial neural network(ANN) model that calculates the priority of the target. Furthermore, we propose a neural network learning algorithm based on the steepest descent method, which is more suitable for target prioritization by combining the conventional gradient descent method. Several simulation results show that the proposed scheme is much more superior to the traditional neural network model from analyzing the training data accuracy and the output priority relevance of the test scenarios.