• Nuclear Engineering and Technology
• /
• v.56 no.3
• /
• pp.993-1001
• /
• 2024

The TANDEM project is a European initiative funded under the EURATOM program. The project started on September 2022 and has a duration of 36 months. TANDEM stands for Small Modular ReacTor for a European sAfe aNd Decarbonized Energy Mix. Small Modular Reactors (SMRs) can be hybridized with other energy sources, storage systems and energy conversion applications to provide electricity, heat and hydrogen. Hybrid energy systems have the potential to strongly contribute to the energy decarbonization targeting carbon-neutrality in Europe by 2050. However, the integration of nuclear reactors, particularly SMRs, in hybrid energy systems, is a new R&D topic to be investigated. In this context, the TANDEM project aims to develop assessments and tools to facilitate the safe and efficient integration of SMRs into low-carbon hybrid energy systems. An open-source "TANDEM" model library of hybrid system components will be developed in Modelica language which, by coupling, will extend the capabilities of existing tools implemented in the project. The project proposes to specifically address the safety issues of SMRs related to their integration into hybrid energy systems, involving specific interactions between SMRs and the rest of the hybrid systems; new initiating events may have to be considered in the safety approach. TANDEM will study two hybrid systems covering the main trends of the European energy policy and market evolution at 2035's horizon: a district heating network and power supply in a large urban area, and an energy hub serving energy conversion systems, including hydrogen production; the energy hub is inspired from a harbor-like infrastructure. TANDEM will provide assessments on SMR safety, hybrid system operationality and techno-economics. Societal considerations will also be encased by analyzing European citizen engagement in SMR technology safety.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.7-21
• /
• 2024

There are various machine learning techniques such as Reinforcement Learning, Deep Learning, Neural Network Learning, and so on. In recent, Large Language Models (LLMs) are popularly used for Generative AI based on Reinforcement Learning. It makes decisions with the most optimal rewards through the fine tuning process in a particular situation. Unfortunately, LLMs can not provide any explanation for how they reach the goal because the training is based on learning of black-box AI. Reinforcement Learning as black-box AI is based on graph-evolving structure for deriving enhanced solution through adjustment by human feedback or reinforced data. In this research, for mutually exclusive decision-making, Mutually Exclusive Learning (MEL) is proposed to provide explanations of the chosen goals that are achieved by a decision on both ends with specified conditions. In MEL, decision-making process is based on the tree-based structure that can provide processes of pruning branches that are used as explanations of how to achieve the goals. The goal can be reached by trade-off among mutually exclusive alternatives according to the specific contextual conditions. Therefore, the tree-based structure is adopted to provide feasible solutions with the explanations based on the pruning branches. The sequence of pruning processes can be used to provide the explanations of the inferences and ways to reach the goals, as Explainable AI (XAI). The learning process is based on the pruning branches according to the multi-dimensional contextual conditions. To deep-dive the search, they are composed of time window to determine the temporal perspective, depth of phases for lookahead and decision criteria to prune branches. The goal depends on the policy of the pruning branches, which can be dynamically changed by configured situation with the specific multi-dimensional contextual conditions at a particular moment. The explanation is represented by the chosen episode among the decision alternatives according to configured situations. In this research, MEL adopts the tree-based learning model to provide explanation for the goal derived with specific conditions. Therefore, as an example of mutually exclusive problems, employment process is proposed to demonstrate the decision-making process of how to reach the goal and explanation by the pruning branches. Finally, further study is discussed to verify the effectiveness of MEL with experiments.

• Journal of Intelligence and Information Systems
• /
• v.24 no.1
• /
• pp.253-266
• /
• 2018

This paper proposes a methodology applying sequence tagging methodology to improve the performance of NER(Named Entity Recognition) used in QA system. In order to retrieve the correct answers stored in the database, it is necessary to switch the user's query into a language of the database such as SQL(Structured Query Language). Then, the computer can recognize the language of the user. This is the process of identifying the class or data name contained in the database. The method of retrieving the words contained in the query in the existing database and recognizing the object does not identify the homophone and the word phrases because it does not consider the context of the user's query. If there are multiple search results, all of them are returned as a result, so there can be many interpretations on the query and the time complexity for the calculation becomes large. To overcome these, this study aims to solve this problem by reflecting the contextual meaning of the query using Bidirectional LSTM-CRF. Also we tried to solve the disadvantages of the neural network model which can't identify the untrained words by using ontology knowledge based feature. Experiments were conducted on the ontology knowledge base of music domain and the performance was evaluated. In order to accurately evaluate the performance of the L-Bidirectional LSTM-CRF proposed in this study, we experimented with converting the words included in the learned query into untrained words in order to test whether the words were included in the database but correctly identified the untrained words. As a result, it was possible to recognize objects considering the context and can recognize the untrained words without re-training the L-Bidirectional LSTM-CRF mode, and it is confirmed that the performance of the object recognition as a whole is improved.

• Journal of the Korean Society for Library and Information Science
• /
• v.57 no.1
• /
• pp.93-114
• /
• 2023

Information extraction can facilitate the intensive analysis of documents by providing semantic triples which consist of named entities and their relations recognized in the texts. However, most of the research so far has been carried out separately for named entity recognition and relation extraction as individual studies, and as a result, the effective performance evaluation of the entire information extraction systems was not performed properly. This paper introduces two models of end-to-end information extraction that can extract various entity names in clinical records and their relationships in the form of semantic triples, namely pipeline and joint models and compares their performances in depth. The pipeline model consists of an entity recognition sub-system based on bidirectional GRU-CRFs and a relation extraction module using multiple encoding scheme, whereas the joint model was implemented with a single bidirectional GRU-CRFs equipped with multi-head labeling method. In the experiments using i2b2/VA 2010, the performance of the pipeline model was 5.5% (F-measure) higher. In addition, through a comparative experiment with existing state-of-the-art systems using large-scale neural language models and manually constructed features, the objective performance level of the end-to-end models implemented in this paper could be identified properly.

• Journal of the Korean Society for Library and Information Science
• /
• v.57 no.2
• /
• pp.435-452
• /
• 2023

This study experimented with automatic classification of subject headings using BERT-based transfer learning model, and analyzed its performance. This study analyzed the classification performance according to the main class of KDC classification and the category type of subject headings. Six datasets were constructed from Korean national bibliographies based on the frequency of the assignments of subject headings, and titles were used as classification features. As a result, classification performance showed values of 0.6059 and 0.5626 on the micro F1 and macro F1 score, respectively, in the dataset (1,539,076 records) containing 3,506 subject headings. In addition, classification performance by the main class of KDC classification showed good performance in the class General works, Natural science, Technology and Language, and low performance in Religion and Arts. As for the performance by the category type of the subject headings, the categories of plant, legal name and product name showed high performance, whereas national treasure/treasure category showed low performance. In a large dataset, the ratio of subject headings that cannot be assigned increases, resulting in a decrease in final performance, and improvement is needed to increase classification performance for low-frequency subject headings.

• Phonetics and Speech Sciences
• /
• v.15 no.4
• /
• pp.109-118
• /
• 2023

Cardiac arrest is a critical medical emergency where immediate response is essential for patient survival. This is especially true for Out-of-Hospital Cardiac Arrest (OHCA), for which the actions of emergency medical services in the early stages significantly impact outcomes. However, in Korea, a challenge arises due to a shortage of dispatcher who handle a large volume of emergency calls. In such situations, the implementation of a machine learning-based OHCA detection program can assist responders and improve patient survival rates. In this study, we address this challenge by developing a machine learning-based OHCA detection program. This program analyzes transcripts of conversations between responders and callers to identify instances of cardiac arrest. The proposed model includes an automatic transcription module for these conversations, a text-based cardiac arrest detection model, and the necessary server and client components for program deployment. Importantly, The experimental results demonstrate the model's effectiveness, achieving a performance score of 79.49% based on the F1 metric and reducing the time needed for cardiac arrest detection by 15 seconds compared to dispatcher. Despite working with a limited dataset, this research highlights the potential of a cardiac arrest detection program as a valuable tool for responders, ultimately enhancing cardiac arrest survival rates.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.11
• /
• pp.64-74
• /
• 1999

A new clock skew optimization for clock routing using link-edge insertion is proposed in this paper. It satisfies the given skew bound and prevent the total wire length from increasing. As the clock skew is the major constraint for high speed synchronous ICs, it must be minimized in order to obtain high performance. But clock skew minimization can increase total wire length, therefore clock routing is performed within the given skew bound which can not induce the malfunction. Clock routing under the specified skew bound can decrease total wire length Not only total wire length and delay time minimization algorithm using merging point relocation method but also clock skew reduction algorithm using link-edge insertion technique between two nodes whose delay difference is large is proposed. The proposed algorithm construct a new clock routing topology which is generalized graph model while previous methods uses only tree-structured routing topology. A new cost function is designed in order to select two nodes which constitute link-edge. Using this cost function, delay difference or clock skew is reduced by connecting two nodes whose delay difference is large and distance difference is short. Furthermore, routing topology construction and wire sizing algorithm is developed to reduce clock delay. The proposed algorithm is implemented in C programming language. From the experimental results, we can get the delay reduction under the given skew bound.

• The KIPS Transactions:PartD
• /
• v.9D no.3
• /
• pp.365-380
• /
• 2002

Temporal data mining, the incorporation of temporal semantics to existing data mining techniques, refers to a set of techniques for discovering implicit and useful temporal knowledge from large quantities of temporal data. Temporal knowledge, expressible in the form of rules, is knowledge with temporal semantics and relationships, such as cyclic pattern, calendric pattern, trends, etc. There are many examples of temporal data, including patient histories, purchaser histories, and web log that it can discover useful temporal knowledge from. Many studies on data mining have been pursued and some of them have involved issues of temporal data mining for discovering temporal knowledge from temporal data, such as sequential pattern, similar time sequence, cyclic and temporal association rules, etc. However, all of the works treated data in database at best as data series in chronological order and did not consider temporal semantics and temporal relationships containing data. In order to solve this problem, we propose a theoretical framework for temporal data mining. This paper surveys the work to date and explores the issues involved in temporal data mining. We then define a model for temporal data mining and suggest SQL-like mining language with ability to express the task of temporal mining and show architecture of temporal mining system.

• Cartoon and Animation Studies
• /
• s.15
• /
• pp.221-235
• /
• 2009

Most of the sounds used in animated films are artificially made. A large number of the sounds used are either actual sound recordings or diversely processed artificial sounds made with professional sound equipments such as synthesizers. One animation episode contains numerous amounts of sounds, resulting in significant sound production costs. These sounds have full potential to be reused in different films or animations, but in reality we fail to do so. This thesis discusses ways these sound sources can be acknowledged as added new values to the present market situation as a usable 'digital content'. The iTunes Music Store is an American Apple company product that is acknowledged as the most successful digital content distribution model at the time being. Its system's sound library has potential for application in the Korean sound industry. In result, this system allows the sound creator to connect directly to the online store and become the initiative content supplier. At the same time, the user can receive a needed content easily at a low price. The most important part in the construction of this system is the search engine, which allows users to search for data in short periods of time. The search engine will have to be made in a new manner that takes into consideration the characteristics of the Korean language. This thesis presents a device incorporating the Wiki System to allow users to search and build their own data bases to share with other users. Using this system as a base, the Korean animation sound library will provide development and growth in the sound source industry as a new digital sound content.

• Journal of KIISE:Software and Applications
• /
• v.37 no.1
• /
• pp.60-73
• /
• 2010

This paper aims to describe KorLexClas 1.5 which provides us with a very large list of Korean numeral classifiers, and with the co-occurring noun categories that select each numeral classifier. Differently from KorLex of other POS, of which the structure depends largely on their reference model (Princeton WordNet), KorLexClas 1.0 and its extended version 1.5 adopt a direct building method. They demand a considerable time and expert knowledge to establish the hierarchies of numeral classifiers and the relationships between lexical items. For the efficiency of construction as well as the reliability of KorLexClas 1.5, we use following processes: (1) to use various language resources while their cross-checking for the selection of classifier candidates; (2) to extend the list of numeral classifiers by using a shallow parsing techniques; (3) to set up the hierarchies of the numeral classifiers based on the previous linguistic studies; and (4) to determine LUB(Least Upper Bound) of the numeral classifiers in KorLexNoun 1.5. The last process provides the open list of the co-occurring nouns for KorLexClas 1.5 with the extensibility. KorLexClas 1.5 is expected to be used in a variety of NLP applications, including MT.