• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.11
• /
• pp.76-88
• /
• 1998

While most of the research on reinforcement learning assumed a discrete control space, many of the real world control problems need to have continuous output. This can be achieved by using continuous mapping functions for the value and action functions of the reinforcement learning architecture. Two questions arise here however. One is what sort of function representation to use and the other is how to determine the amount of noise for search in action space. The ubiquitous neural network is used here to learn the value and policy functions. Next, the reinforcement predictor that is intended to predict the next reinforcement is introduced that also determines the amount of noise to add to the controller output. The proposed reinforcement learning architecture is found to have a sound on-line learning control performance especially at high-speed road following of high curvature road. Both computer simulation and actual experiments on a test vehicle have been performed and their efficiency and effectiveness has been verified.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.26 no.6
• /
• pp.601-605
• /
• 2020

In large ships with complex structures, it is difficult to locate workers. In particular, it is not easy to detect when a worker falls down, making it difficult to respond quickly. Thus, research is being conducted to detect fallen workers using a camera or by attaching a device to the body. Existing image-based fall detection systems have been designed to detect a person's body parts; hence, it is difficult to detect them in various ships and postures. In this study, the entire fall area was extracted and deep learning was used to detect the fallen shipworker based on the image. The data necessary for learning were obtained by recording falling states at the shipyard. The amount of learning data was augmented by flipping, resizing, and rotating the image. Performance evaluation was conducted with precision, reproducibility, accuracy, and a low error rate. The larger the amount of data, the better the precision. In the future, reinforcing various data is expected to improve the effectiveness of camera-based fall detection models, and thus improve safety.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.1
• /
• pp.37-46
• /
• 2019

The energy of the building is influenced by the user 's activity due to the population, society, and economic characteristics of the building user. In order to obtain accurate energy information, the difference in the amount of energy consumption by the activities and characteristics of building users should be identified. The purpose of the study is to identify the difference in the amount of energy consumption by the user's activities in the same building, and to analyse the relationship between user's activities and demographic, social and economic characteristics. For research, energy simulation is performed based on actual user activity schedule. The results of the simulation were clustered by using K-Means clustering, a machine learning technique. As a result, four types of users were derived based on the amount of energy consumption. The more energy used in a cluster, the lower the user's income level and older. The longer a user's indoor activity times, the higher the energy use, and these activities relate to the user's characteristics. There is more than twice the difference between the group that uses the least energy consumption and the group that uses the most energy consumption.

• Asia pacific journal of information systems
• /
• v.29 no.4
• /
• pp.789-816
• /
• 2019

Text classification is a challenging task, especially when dealing with a huge amount of text data. The performance of a classification model can be varied depending on what type of words contained in the document corpus and what type of features generated for classification. Aside from proposing a new modified version of the existing algorithm or creating a new algorithm, we attempt to modify the use of data. The classifier performance is usually affected by the quality of learning data as the classifier is built based on these training data. We assume that the data from different domains might have different characteristics of noise, which can be utilized in the process of learning the classifier. Therefore, we attempt to enhance the robustness of the classifier by injecting the heterogeneous data artificially into the learning process in order to improve the classification accuracy. Semi-supervised approach was applied for utilizing the heterogeneous data in the process of learning the document classifier. However, the performance of document classifier might be degraded by the unlabeled data. Therefore, we further proposed an algorithm to extract only the documents that contribute to the accuracy improvement of the classifier.

• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.18-28
• /
• 2017

Recently, there are many companies that use systems based on artificial intelligence. The accuracy of artificial intelligence depends on the amount of learning data and the appropriate algorithm. However, it is not easy to obtain learning data with a large number of entity. Less data set have large generalization errors due to overfitting. In order to minimize this generalization error, this study proposed DGA which can expect relatively high accuracy even though data with a less data set is applied to machine learning based genetic algorithm to deep learning based dropout. The idea of this paper is to determine the active state of the nodes. Using Gradient about loss function, A new fitness function is defined. Proposed Algorithm DGA is supplementing stochastic inconsistency about Dropout. Also DGA solved problem by the complexity of the fitness function and expression range of the model about Genetic Algorithm As a result of experiments using MNIST data proposed algorithm accuracy is 75.3%. Using only Dropout algorithm accuracy is 41.4%. It is shown that DGA is better than using only dropout.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.6
• /
• pp.287-294
• /
• 2019

With the automation of production lines in the manufacturing industry, the importance of real-time fault diagnosis of facility is increasing. In this paper, we propose a fault diagnosis algorithm of LM (Linear Motion)-guide based on deep learning using vibration signals. Generally, in order to guarantee the performance of the deep learning, it is necessary to have a sufficient amount of data, but in a manufacturing industry, it is often difficult to obtain enough data due to physical and time constraints. To solve this problem, we propose a convolutional neural networks (CNN) model based on transfer learning. In addition, the spectrogram image is input to the CNN to reflect the frequency characteristic of the vibration signals with time. The performance of fault diagnosis according to various load condition and transfer learning method was compared and evaluated by experiments. The results showed that the proposed algorithm exhibited an excellent performance.

• Journal of radiological science and technology
• /
• v.42 no.1
• /
• pp.9-17
• /
• 2019

Although digital mammography is a representative method for breast cancer detection. It has a limitation in detecting and classifying breast tumor due to superimposed structures. Machine learning, which is a part of artificial intelligence fields, is a method for analysing a large amount of data using complex algorithms, recognizing patterns and making prediction. In this study, we proposed a technique to improve the diagnostic accuracy of energy-selective mammography by training data using the machine learning algorithm and using dual-energy measurements. A dual-energy images obtained from a photon-counting detector were used for the input data of machine learning algorithms, and we analyzed the accuracy of predicted tumor thickness for verifying the machine learning algorithms. The results showed that the classification accuracy of tumor thickness was above 95% and was improved with an increase of imput data. Therefore, we expect that the diagnostic accuracy of energy-selective mammography can be improved by using machine learning.

• Journal of Korean Critical Care Nursing
• /
• v.7 no.2
• /
• pp.1-13
• /
• 2014

Purpose: The purpose of this study was to determine the difference in reported discharge learning needs between nurses and liver transplantation (LT) patients. Methods: The participants of this study were 40 patients discharged after LT at P University Hospital in Y City and 42 nurses in intensive care units and the ward. The data were collected for two months from December 1, 2012, to January 31, 2013, and were analyzed using descriptive statistics, Student's t-test and analysis of variance (ANOVA). Results: Patients earning a low income (p=.041), having no experience of hospitalization after LT (p=.023), and receiving information about LT from nurses (p=.003) indicated higher discharge learning needs. Among the items evaluated regarding discharge learning needs, "rejection symptoms or signs" were regarded to be more important by nurses than LT patients (p=.038). However, "management of other diseases after LT" (p=.003), "risk of recurrence" (p=.001), "food choices" (p<.001), "obesity prevention" (p=.020), "amount of exercise" (p=.007), and "ways to receive financial help"(p=.033), were thought to be more important by LT patients than nurses. Conclusion: There exist differences between LT patients and nurses with respect to their perceptions of LT discharge learning needs. Therefore, an individualized education program reflecting patients' conditions and learning needs rather than providing information uniformly needs to be developed.

• Korean Journal of Artificial Intelligence
• /
• v.11 no.1
• /
• pp.17-24
• /
• 2023

With the continuous growth in the amount of data collected and analyzed, deep learning has become increasingly popular for extracting meaningful insights from various fields. However, hardware limitations pose a challenge for achieving meaningful results with limited data. To address this challenge, this paper proposes an algorithm that leverages the characteristics of convolutional neural networks (CNNs) to reduce the size of image datasets by 20% through smoothing and shrinking the size of images using color elements. The proposed algorithm reduces the learning time and, as a result, the computational load on hardware. The experiments conducted in this study show that the proposed method achieves effective learning with similar or slightly higher accuracy than the original dataset while reducing computational and time costs. This color-centric dataset construction method using image smoothing techniques can lead to more efficient learning on CNNs. This method can be applied in various applications, such as image classification and recognition, and can contribute to more efficient and cost-effective deep learning. This paper presents a promising approach to reducing the computational load and time costs associated with deep learning and provides meaningful results with limited data, enabling them to apply deep learning to a broader range of applications.

• The Journal of Korean Physical Therapy
• /
• v.31 no.4
• /
• pp.212-215
• /
• 2019

Purpose: This study examined whether there is a difference in motor learning through short-term repetitive movement practice in stroke survivors with a unilateral brain injury compared to normal elderly participants. Methods: Twenty-six subjects who were divided into a stroke group (n=13) or sex-aged matched normal elder group (n=13) participated in this study. To evaluate the effects of motor learning, the participants conducted a tracking task for visuomotor coordination. The accuracy index was calculated for each trial. Both groups received repetitive tracking task training of metacarpophalangeal joint for 50 trials. The stroke group performed a tracking task in the upper extremity insi-lesional to the damaged hemisphere, and the normal elder group performed the upper extremity matched for the same side. Results: Two-way repetitive ANOVA revealed a significant difference in the interactions ($time{\times}group$) and time effects. These results indicated that the motor skill improved in both the stroke and normal elder group with a tracking task. On the other hand, the stroke group showed lesser motor learning skill than the normal elder group, in comparison with the amount of motor learning improvement. Conclusion: These results provide novel evidence that stroke survivors with unilateral brain damage might have difficulty in performing ipsilateral movement as well as in motor learning with the ipsilateral upper limb, compared to normal elderly participants.