• The Korean Journal of Applied Statistics
• /
• v.34 no.5
• /
• pp.745-760
• /
• 2021

Prediction of a stock price has been a subject of interest for a long time in financial markets, and thus, many studies have been conducted in various directions. As the efficient market hypothesis introduced in the 1970s acquired supports, it came to be the majority opinion that it was impossible to predict stock prices. However, recent advances in predictive models have led to new attempts to predict the future prices. Here, we summarize past studies on the price prediction by evaluation measures, and predict the direction of stock prices of Samsung Electronics, LG Chem, and NAVER by applying various machine learning models. In addition to widely used technical indicator variables, accounting indicators such as Price Earning Ratio and Price Book-value Ratio and outputs of the hidden Markov Model are used as predictors. From the results of our analysis, we conclude that no models show significantly better accuracy and it is not possible to predict the direction of stock prices with models used. Considering that the models with extra predictors show relatively high test accuracy, we may expect the possibility of a meaningful improvement in prediction accuracy if proper variables that reflect the opinions and sentiments of investors would be utilized.

• Journal of Internet Computing and Services
• /
• v.22 no.1
• /
• pp.23-30
• /
• 2021

This paper proposes a noise-tolerant image classification system using multiple autoencoders. The development of deep learning technology has dramatically improved the performance of image classifiers. However, if the images are contaminated by noise, the performance degrades rapidly. Noise added to the image is inevitably generated in the process of obtaining and transmitting the image. Therefore, in order to use the classifier in a real environment, we have to deal with the noise. On the other hand, the autoencoder is an artificial neural network model that is trained to have similar input and output values. If the input data is similar to the training data, the error between the input data and output data of the autoencoder will be small. However, if the input data is not similar to the training data, the error will be large. The proposed system uses the relationship between the input data and the output data of the autoencoder, and it has two phases to classify the images. In the first phase, the classes with the highest likelihood of classification are selected and subject to the procedure again in the second phase. For the performance analysis of the proposed system, classification accuracy was tested on a Gaussian noise-contaminated MNIST dataset. As a result of the experiment, it was confirmed that the proposed system in the noisy environment has higher accuracy than the CNN-based classification technique.

• Therapeutic Science for Rehabilitation
• /
• v.10 no.2
• /
• pp.37-51
• /
• 2021

Objective : The purpose of this study was to analyze non-invasive treatments and drooling assessment methods in children with cerebral palsy and developmental disabilities, who drool. Methods : This study searched two hundred papers published in 2005-2019. Forty-four papers were selected based on their abstract and title, and ten papers were finally selected following a secondary search. Results : The PEDro Scale of the selected papers was high with an average of seven points. As a result of analyzing the overall trends, the study participants were primarily patients with cerebral palsy, and recently, the therapeutic intervention of oral sensory exercise was more actively studied than behavioral modification. Studies of behavioral modification and oral sensory exercise intervention methods were found to have differences in participant age and, cognitive level, number of participants, research design, treatment time, and duration. Studies to confirming the frequency and severity of the drooling measurement method were found to be the main factor. Conclusion : This study analyzed typical behavioral modification and oral sensory exercise interventions as examples of non-invasive therapeutic interventions for children with cerebral palsy and developmental disabilities and provided information to help select appropriate therapeutic intervention methods when planning non-invasive therapy using behavioral modification and oral sensory exercise therapy.

• The Journal of Korean Academy of Sensory Integration
• /
• v.20 no.1
• /
• pp.55-72
• /
• 2022

Objective : The purpose of this study was to identify the status and trends of Korean child occupational therapy intervention studies according to the International Classification of Functioning, Disability, and Health, Children and Youth Version (ICF-CY). Methods : In this research, 47 studies on occupational therapy interventions for children that were published between January 2017 and December 2021 in the Journal of Occupational Therapy, registered in the Korea Citation Index, and analyzed the classification of the study type and evidence level to understand the trends. Moreover, intervention objectives and approaches were analyzed on the basis of the ICF-CY. Results : The outcomes of the analysis of the articles published in the Journal of Occupational Therapy were as follows: (1) Level IV was the highest evidence level (53.19%). (2) Among the studies, most (53.7%) included school-age children as subjects. Autism spectrum disorders and developmental delays were the most common diagnoses (14.8%). (3) As for the purpose of intervention according to ICF-CY, activity and participation factors were the most common (48.94%), and a sensory approach was frequently used. Conclusion : This study reviewed articles on occupational therapies for children that were published in the Journal of Occupational Therapy to understand the trends in occupational therapy interventions for children in South Korea. For the development of occupational therapies for children in the future, more qualitative research types and studies on various intervention approaches are needed.

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.27 no.2
• /
• pp.49-70
• /
• 2022

Sea Surface Temperature (SST), one of the ocean features, has a significant impact on climate, marine ecosystem and human activities. Therefore, SST prediction has been always an important issue. Recently, deep learning has drawn much attentions, since it can predict SST by training past SST patterns. Compared to the numerical simulations, deep learning model is highly efficient, since it can estimate nonlinear relationships between input data. With the recent development of Graphics Processing Unit (GPU) in computer, large amounts of data can be calculated repeatedly and rapidly. In this study, Short-term SST will be predicted through Convolutional Neural Network (CNN)-based U-Net that can handle spatiotemporal data concurrently and overcome the drawbacks of previously existing deep learning-based models. The SST prediction performance depends on the seasonal and interannual SST variabilities around the southern coast of Korea. The predicted SST has a wide range of variance during spring and summer, while it has small range of variance during fall and winter. A wide range of variance also has a significant correlation with the change of the Pacific Decadal Oscillation (PDO) index. These results are found to be affected by the intensity of the seasonal and PDO-related interannual SST fronts and their intensity variations along the southern Korean seas. This study implies that the SST prediction performance using the developed deep learning model can be significantly varied by seasonal and interannual variabilities in SST.

• Journal of Marine Life Science
• /
• v.5 no.2
• /
• pp.58-63
• /
• 2020

The ascidian Halocynthia aurantium (sea peach), which belongs to the phylum Chordata, is thought to be a valuable organism of aquaculture like H. roretzi (sea pineapple), but its biological characteristics such as development and ecology are not well known. In this study, in order to obtain basic data for H. aurantium farming, the development processes of H. aurantium inhabiting the east coast of Gangwon-do were investigated and compared with those of H. roretzi, a related species. As a result, the morphology and developmental stages of the fertilized eggs, embryos and larvae of H. aurantium were very similar to those of H. roretzi. Fertilized eggs of H. aurantium took about 42.1 hours to hatch at 11℃, almost similar to 40.9 hours of H. roretzi. The time required for larvae to metamorphose into juveniles after hatching was very similar between the two species. The hatched larvae of the two species became juveniles with oral and atrial siphons after 23 days at 11℃. Both types of embryos developed slowly in seawater at low temperatures and rapidly developed at high temperatures. Fertilized eggs of H. aurantium hatched in an average of 62.3 hours at 9℃, 42.1 hours at 11℃, and 36.3 hours at 13℃, whereas those of H. roretzi hatched in an average of 60.4 hours, 40.9 hours, and 35.2 hours. Most of H. aurantium embryos did not develop normally above 15℃, so it is thought that attention is needed in the seed production processes.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.5
• /
• pp.697-703
• /
• 2022

Due to the recent economic downturn caused by Covid-19 and the unstable international situation, many investors are choosing the derivatives market as a means of investment. However, the derivatives market has a greater risk than the stock market, and research on the market of market participants is insufficient. Recently, with the development of artificial intelligence, machine learning has been widely used in the derivatives market. In this paper, reinforcement learning, one of the machine learning techniques, is applied to analyze the scalping technique that trades futures in minutes. The data set consists of 21 attributes using the closing price, moving average line, and Bollinger band indicators of 1 minute and 3 minute data for 6 months by selecting 4 products among futures products traded at trading firm. In the experiment, DNN artificial neural network model and three reinforcement learning algorithms, namely, DQN (Deep Q-Network), A2C (Advantage Actor Critic), and A3C (Asynchronous A2C) were used, and they were trained and verified through learning data set and test data set. For scalping, the agent chooses one of the actions of buying and selling, and the ratio of the portfolio value according to the action result is rewarded. Experiment results show that the energy sector products such as Heating Oil and Crude Oil yield relatively high cumulative returns compared to the index sector products such as Mini Russell 2000 and Hang Seng Index.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.1
• /
• pp.41-49
• /
• 2023

A convolutional neural network (CNN) is a representative algorithm for implementing artificial neural networks. CNNs have improved on the issues of rapid increase in calculation amount and low object classification rates, which are associated with a conventional multi-layered fully-connected neural network (FNN). However, because of the rapid development of IT devices, the maximum resolution of images captured by current smartphone and tablet cameras has reached 108 million pixels (MP). Specifically, a traditional CNN algorithm requires a significant cost and time to learn and process simple, high-resolution images. Therefore, this study proposes an improved CNN algorithm for implementing an object classification learning model for simple, high-resolution images. The proposed method alters the adjacency matrix value of the pooling layer's max pooling operation for the CNN algorithm to reduce the high-resolution image learning model's creation time. This study implemented a learning model capable of processing 4, 8, and 12 MP high-resolution images for each altered matrix value. The performance evaluation result showed that the creation time of the learning model implemented with the proposed algorithm decreased by 36.26% for 12 MP images. Compared to the conventional model, the proposed learning model's object recognition accuracy and loss rate were less than 1%, which is within the acceptable error range. Practical verification is necessary through future studies by implementing a learning model with more varied image types and a larger amount of image data than those used in this study.

• Therapeutic Science for Rehabilitation
• /
• v.12 no.2
• /
• pp.25-42
• /
• 2023

Objective : The purpose of this study was to identify the characteristics of a single-subject research study and qualitative levels in which occupational therapy interventions were performed on children with cerebral palsy. Methods : This study targeted papers, published in Research Information Sharing Service (RISS), National Digital Science Library (NDSL), Koreanstudies Information Service System (KISS), and E-article from May 20 to 29, 2022. The search terms were 'cerebral palsy' AND 'single subject research design' OR 'individual subject study'. Eleven papers, were finally selected and analyzed. Results : Most of the studies were medium in methodological quality, and the subjects were pre-school age and spastic hemiplegia. Among the single-subject designs, intervention-removal designs were the most common, and among them, ABA designs were the most common. Interventions included assistive devices, constraint-induced therapy, neurodevelopmental therapy, and sensory integration therapy were 2, and upper extremity exercise, interactive metronome, and CO-OP were 1. Dependent variables were measured with 2 to 4 measurement tools, Significant improvements were found in postural control ability, gait and balance, hand function, and upper extremity function. Conclusion : This study confirmed that it is helpful to apply cerebral palsy occupational therapy by presenting the characteristics of cerebral palsy, intervention sessions and effects, measurement tools and methodological quality levels.

• Clinical and Experimental Pediatrics
• /
• v.52 no.11
• /
• pp.1228-1233
• /
• 2009

Purpose : This study aimed to evaluate whether a shorter time from the arrival at a hospital to delivery is related to the occurrence of cerebral palsy in premature infants of less than 34 weeks of gestational age. Methods : We studied 142 newborns of less than 34 weeks of gestational age. The time from the arrival at the hospital to delivery was measured. The correlation between the time required for delivery and the occurrence of cerebral palsy was elucidated by diagnosing cerebral palsy in neonates using the Korean Infant Development Screening Test and neurological examination. Results : Preliminary result suggested that a shorter time from hospital arrival to delivery was related to a lower development score for gross motor activity and to a higher frequency of cerebral palsy occurrence. Moreover, it was responsible for a tendency of obtaining lower Apgar scores at 1 and 5 minutes. The shorter delivery time was associated with a higher probability of respiratory distress syndrome (RDS) occurrence when the length of delivery time was less than 6 hours and there was a higher probability of a shorter gestation period. However, the multifactor analysis revealed that there was little impact of delivery time on the occurrence of cerebral palsy. Conclusions : The length of hospital arrival time to delivery did not significantly influence the occurrence of cerebral palsy in premature infants of less than 34 weeks of gestational age.