• Electronics and Telecommunications Trends
• /
• v.34 no.6
• /
• pp.100-107
• /
• 2019

Recently, reinforcement learning (RL) has expanded from the research phase of the virtual simulation environment to a wide range of applications, such as autonomous driving, natural language processing, recommendation systems, and disease diagnosis. However, RL is less likely to be used in these complex real-world environments. In contrast, inverse reinforcement learning (IRL) can obtain optimal policies in various situations; furthermore, it can use expert demonstration data to achieve its target task. In particular, IRL is expected to be a key technology for artificial general intelligence research that can successfully perform human intellectual tasks. In this report, we briefly summarize various IRL techniques and research directions.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.3
• /
• pp.421-428
• /
• 2021

Identifying which international maritime legal instruments are mandatory or recommendatory is complicated task even for maritime regulatory bodies. Although International Maritime Organization (IMO) had tried to ease the complexity by adopting guidelines on uniform wordings for making reference to other instruments in IMO parent conventions, there has still been some confusion identifying the mandatory status of IMO instruments. The aim of this study was to map out a step-based guideline to resolve the complexity of the mandatory status of IMO instruments to the maximum extent possible. This study reviewed the history of IMO rule-making process to find the root cause of the problem, then analyzed the approaches of regulatory enforcement bodies to check the practices. In conclusion, readers are directed to find such information as to legal status of IMO instruments and an improvement is proposed to enhance the transparency of information sharing for maritime industry to make better informed decisions.

• Korea Trade Review
• /
• v.47 no.4
• /
• pp.15-33
• /
• 2022

This study is to investigate multilateral environmental agreements,mainly UNFCCC on the primary export industry of Korea and to make a policy recommendation. Mostly literature reviews are focused on the traditional multilateral environmental agreements and the for the most part analysis are conducted prior to the Paris agreement. The result of survey indicates that many companies have not yet felt burden on their business due to UNFCCC(decarbonization) and have monitored the related policies. But the companies ask the government for strong incentives. The paper implies that enforcing strong government incentives, upgrading usage of the nuclear power, improving the related government legislation, setting up the special task force team with government and private sectors are needed.

• Nuclear Engineering and Technology
• /
• v.56 no.7
• /
• pp.2516-2523
• /
• 2024

This study aims to validate the dosimetric characteristics of Low Dose Rate (LDR) I-125 source Geant4-based Monte Carlo code. According to the recommendation of the American Association of Physicists in Medicine (AAPM) task group report (TG-43), the dosimetric parameters of a new brachytherapy source should be verified either experimentally or theoretically before clinical procedures. The simulation studies are very important since this procedure delivers a high dose of radiation to the tumor with only a minimal dose affecting the surrounding tissues. GEANT4 Monte Carlo simulation toolkit associated brachytherapy example was modified, adapted and several updated techniques have been developed to facilitate and smooth radiotherapy techniques. The great concordance of the current study results with the consensus data and with the results of other MC based studies is promising. It implies that Geant4-based Monte Carlo simulation has the potential to be used as a reliable and standard simulation code in the field of brachytherapy for verification and treatment planning purposes.

• Journal of Korea Multimedia Society
• /
• v.5 no.3
• /
• pp.239-248
• /
• 2002

International Standard Organizations such as ITU(International Telecommunication Union) and IETF (Internet Engineering Task Force) are proceeding standardization for various applications and protocols to provide video-conference and multimedia conference services on a variety of networks. Remote device control among these protocols is provided with various capabilities as well as device control to multimedia conference. This protocol for remote device control is standardizing as H.282 recommendation which is specified as core service for the configuration and control of remote device to multimedia conference. The H.282 recommendation does not specify the use of a particular transport protocol. That is, T.120 multimedia conference uses T.136 and H.323 video conference uses H.283 for the transport of H.282 protocol. The introduced system in this paper is based on H.282 and is implemented to be capable of remote device control within the framework of H.323 using H.283. Also, it is shown that a variety of services in the specification of the standard are satisfied through experiments.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.12
• /
• pp.1809-1816
• /
• 2021

In order to identify the latest research trends using data related to national R&D projects and to produce and utilize meaningful information, the application of automatic classification technology was also required in the national R&D information service, so we conducted research to automatically classify and recommend research field. About 450,000 cases of national R&D project data from 2013 to 2020 were collected and used for learning and evaluation. A model was selected after data pre-processing, analysis, and performance analysis for valid data among collected data. The performance of Word2vec, GloVe, and fastText was compared for the purpose of deriving the optimal model combination. As a result of the experiment, the accuracy of only the subcategories used as essential items of task information is 90.11%. This model is expected to be applicable to the automatic classification study of other classification systems with a hierarchical structure similar to that of the national science and technology standard classification research field.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.5
• /
• pp.1413-1432
• /
• 2023

Recommender systems aim to recommend items to users by taking into account their probable interests. This study focuses on creating a model that utilizes multiple sources of information about users and items by employing a multimodality approach. The study addresses the task of how to gather information from different sources (modalities) and transform them into a uniform format, resulting in a multi-modal feature description for users and items. This work also aims to transform and represent the features extracted from different modalities so that the information is in a compatible format for integration and contains important, useful information for the prediction model. To achieve this goal, we propose a novel multi-modal recommendation model, which involves extracting latent features of users and items from a utility matrix using matrix factorization techniques. Various transformation techniques are utilized to extract features from other sources of information such as user reviews, item descriptions, and item categories. We also proposed the use of Principal Component Analysis (PCA) and Feature Selection techniques to reduce the data dimension and extract important features as well as remove noisy features to increase the accuracy of the model. We conducted several different experimental models based on different subsets of modalities on the MovieLens and Amazon sub-category datasets. According to the experimental results, the proposed model significantly enhances the accuracy of recommendations when compared to SVD, which is acknowledged as one of the most effective models for recommender systems. Specifically, the proposed model reduces the RMSE by a range of 4.8% to 21.43% and increases the Precision by a range of 2.07% to 26.49% for the Amazon datasets. Similarly, for the MovieLens dataset, the proposed model reduces the RMSE by 45.61% and increases the Precision by 14.06%. Additionally, the experimental results on both datasets demonstrate that combining information from multiple modalities in the proposed model leads to superior outcomes compared to relying on a single type of information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1773-1793
• /
• 2023

Automated machine learning, often referred to as "AutoML," is the process of automating the time-consuming and iterative procedures that are associated with the building of machine learning models. There have been significant contributions in this area across a number of different stages of accomplishing a data-mining task, including model selection, hyper-parameter optimization, and preprocessing method selection. Among them, preprocessing method selection is a relatively new and fast growing research area. The current work is focused on the recommendation of preprocessing methods, i.e., feature subset selection (FSS) algorithms. One limitation in the existing studies regarding FSS algorithm recommendation is the use of a single learner for meta-modeling, which restricts its capabilities in the metamodeling. Moreover, the meta-modeling in the existing studies is typically based on a single group of data characterization measures (DCMs). Nonetheless, there are a number of complementary DCM groups, and their combination will allow them to leverage their diversity, resulting in improved meta-modeling. This study aims to address these limitations by proposing an architecture for preprocess method selection that uses ensemble learning for meta-modeling, namely AutoFE-Sel. To evaluate the proposed method, we performed an extensive experimental evaluation involving 8 FSS algorithms, 3 groups of DCMs, and 125 datasets. Results show that the proposed method achieves better performance compared to three baseline methods. The proposed architecture can also be easily extended to other preprocessing method selections, e.g., noise-filter selection and imbalance handling method selection.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.8
• /
• pp.2780-2792
• /
• 2010

In this study, we examine what components of e-learning environment affect e-learners' satisfaction. We focus on the task based approach to perceived website complexity(PWC). We study about the role of e-tutor using the internet, telephone, text message and e-mail etc. To test our model, we collected 235 data from online learners of Korea Culture & Content Agency using survey method. The research was conducted by SPSS15.0. Our results show that the relationship between PWC and e-learner satisfaction was negative. The rules of e-tutor are supporting e-learning service and facilitating recommendation intention. This study provides implications to design future e-learning service, understand user's herd behavior and evaluate learning process developed.

• Industrial Engineering and Management Systems
• /
• v.6 no.1
• /
• pp.40-48
• /
• 2007

This study investigated the optimum positions and warning foreperiod for auditory signals with an experiment on spatial stimulus-response (S-R) compatibility effects. The auditory signals were presented at the front-right, front-left, rear-right, and rear-left positions from the subjects, whose reaction times and accuracies at different spatial mapping conditions were examined. The results showed a significant spatial stimulus-response compatibility effect in which faster and more accurate responses were obtained in the transversely and longitudinally compatible condition while the worst performance was found when spatial stimulus-response compatibility did not exist in either orientation. It was also shown that the transverse compatibility effect was found significantly stronger than the longitudinal compatibility effect. The effect of signal position was found significant and post hoc test suggested that the emergent warning alarm should be placed on the front-right position for right-handed users. The warning foreperiod prior to the signal presentation was shown to influence reaction time and a warning foreperiod of 3 s is found optimal for the 2-choice auditory reaction task.