• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.27-40
• /
• 2022

Medical AI, which has lately made significant advances, is playing a vital role, such as assisting clinicians with diagnosis and decision-making. The field of chest X-rays, in particular, is attracting a lot of attention since it is important for accessibility and identification of chest diseases, as well as the current COVID-19 pandemic. However, despite the vast amount of data, there remains a limit to developing an effective AI model due to a lack of labeled data. A research that used federated learning on chest X-ray data to lessen this difficulty has emerged, although it still has the following limitations. 1) It does not consider the problems that may occur in the Non-IID environment. 2) Even in the federated learning environment, there is still a shortage of labeled data of clients. We propose a method to solve the above problems by using the self-supervised learning model as a global model of federated learning. To that aim, we investigate a self-supervised learning methods suited for federated learning using chest X-ray data and demonstrate the benefits of adopting the self-supervised learning model for federated learning.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.835-843
• /
• 2023

Given the structure of XML data, path and tree pattern matching algorithms play an important role in XML query processing. To facilitate decisions or relationships between nodes, nodes in an XML tree are typically labeled in a way that can quickly establish an ancestor-descendant on relationship between two nodes. However, these techniques have the disadvantage of re-labeling existing nodes or recalculating certain values if insertion occurs due to sequential updates. Therefore, in current labeling techniques, the cost of updating labels is very high. In this paper, we propose a new labeling technique called Fast XML encoding, which supports the update of order-sensitive XML documents without re-labeling or recalculation. It also controls the length of the label by reusing deleted labels at the same location in the XML tree. The proposed reuse algorithm can reduce the length of the label when all deleted labels are inserted in the same location. The proposed technique in the experimental results can efficiently handle order-sensitive queries and updates.

• Journal of KIISE:Information Networking
• /
• v.28 no.2
• /
• pp.251-261
• /
• 2001

The traffic on the Internet has been growing exponentially for some time. This growth is beginning to stress the current-day routers. However, switching technology offers much higher performance. So the label switching network which combines IP routing with switching technology, is emerged. EspeciaJJy in the data driven label switching, flow classification and cache table management are needed. Flow classification is to classify packets into switching and non-switching packets, and cache table management is to maintain the cache table which contains information for flow classification and label switching. However, the cache table management affects the performance of label switching network considerably as well as flowclassification because the bigger cache table makes more packet switched and maintains setup cost lower, but cache is restricted by local router resources. For that reason, there is need to study the cache replacement scheme for the efficient cache table management with the Internet traffic characterized by user. So in this paper, we propose several cache replacement schemes for label switching network. First, without the limitation at switching capacity in the router. we introduce FIFO(First In First Out). LFC(Least Flow Count), LRU(Least Recently Used! scheme and propose priority LRU, weighted priority LRU scheme. Second, with the limitation at switching capacity in the router, we introduce LFC-LFC, LFC-LRU, LRU-LFC, LRU-LRU scheme and propose LRU-weighted LRU scheme. Without limitation, weighted priority LRU scheme and with limitation, LRU-weighted LRU scheme showed best performance in this paper.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.300-304
• /
• 2016

본 논문에서는 한국어 의미역 결정을 순차열 분류 문제(Sequence Labeling Problem)가 아닌 순차열 변환 문제(Sequence-to-Sequence Learning)로 접근하였고, 구문 분석 단계와 자질 설계가 필요 없는 End-to-end 방식으로 연구를 진행하였다. 음절 단위의 RNN Search 모델을 사용하여 음절 단위로 입력된 문장을 의미역이 달린 어절들로 변환하였다. 또한 순차열 변환 문제의 성능을 높이기 위해 연구된 인풋-피딩(Input-feeding) 기술과 카피넷(CopyNet) 기술을 한국어 의미역 결정에 적용하였다. 실험 결과, Korean PropBank 데이터에서 79.42%의 레이블 단위 f1-score, 71.58%의 어절 단위 f1-score를 보였다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.300-304
• /
• 2016

본 논문에서는 한국어 의미역 결정을 순차열 분류 문제(Sequence Labeling Problem)가 아닌 순차열 변환 문제(Sequence-to-Sequence Learning)로 접근하였고, 구문 분석 단계와 자질 설계가 필요 없는 End-to-end 방식으로 연구를 진행하였다. 음절 단위의 RNN Search 모델을 사용하여 음절 단위로 입력된 문장을 의미역이 달린 어절들로 변환하였다. 또한 순차열 변환 문제의 성능을 높이기 위해 연구된 인풋-피딩(Input-feeding) 기술과 카피넷(CopyNet) 기술을 한국어 의미역 결정에 적용하였다. 실험 결과, Korean PropBank 데이터에서 79.42%의 레이블 단위 f1-score, 71.58%의 어절 단위 f1-score를 보였다.

• Journal of Navigation and Port Research
• /
• v.47 no.6
• /
• pp.358-366
• /
• 2023

This study aimed to provide a solution for improving ship collision alert of the 'accident vulnerable ship monitoring service' among the 'intelligent marine traffic information system' services of the Ministry of Oceans and Fisheries. The current ship collision alert uses a supervised learning (SL) model with survey labels based on large ship-oriented data and its operators. Consequently, the small ship data and the operator's opinion are not reflected in the current collision-supervised learning model, and the effect is insufficient because the alarm is provided from a longer distance than the small ship operator feels. In addition, the supervised learning (SL) method requires a large number of labeled data, and the labeling process requires a lot of resources and time. To overcome these limitations, in this paper, the classification model of collision alerts for small ships using unlabeled data with the semi-supervised learning (SSL) algorithms (Label Propagation and TabNet) was studied. Results of real-time experiments on small ship operators using the classification model of collision alerts showed that the satisfaction of operators increased.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.37-53
• /
• 2023

The AI based intelligent command and control system can automatically analyzes the properties of intricate battlefield information and tactical data. In addition, commanders can receive situation analysis results and battlefield awareness through the system to support decision-making. It is necessary to build a battlefield situation analysis dataset similar to the actual battlefield situation for learning AI in order to provide decision-making support to commanders. In this paper, we explain the next step of the dataset construction method of the existing previous research, 'A Virtual Battlefield Situation Dataset Generation for Battlefield Analysis based on Artificial Intelligence'. We proposed a method to build the dataset required for the final battlefield situation analysis results to support the commander's decision-making and recognize the future battlefield. We developed 'Dataset Generator SW', a software tool to build a learning dataset for battlefield situation analysis, and used the SW tool to perform data labeling. The constructed dataset was input into the Siamese Network model. Then, the output results were inferred to verify the dataset construction method using a post-processing ranking algorithm.

• Annual Conference on Human and Language Technology
• /
• 2018.10a
• /
• pp.373-377
• /
• 2018

판례는 재판에 대한 선례로, 법적 결정에 대한 근거가 되는 핵심 단서 중 하나이다. 본 연구에서는 채권회수를 예측하는 서비스 구축을 위한 단서를 추출하기 위해 채권 회수 판례를 수집하여 이를 분석한다. 먼저 채권 회수 판례에 대한 기초 분석을 위하여, 채권 회수 사례와 비회수 사례를 각 20건씩 수집하여 분석하였으며, 이후 대법원 및 법률 지식베이스의 채권 관련 판례 12,457건을 수집하고 채권 회수 여부에 따라 가공하였다. 채권 회수 사례와 비회수 사례를 분류하기 위한 판례 내의 패턴을 분석하여 레이블링하고, 이를 자동 분류할 수 있는 Bidirectional LSTM 기반 심층학습 모델을 구성하여 학습하였다. 채권 관련 판례 가공 기준에 따라 네 가지의 데이터 셋을 구성하였으며, 각 데이터셋을 8:2의 비율로 나누어 실험한 결과, 검증 데이터에 대하여 F1 점수 89.82%의 우수한 성능을 보였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.05a
• /
• pp.422-425
• /
• 2009

RFID (Radio Frequency IDentificaiton)는 태그, 레이블, 카드 등에 저장된 데이터를 무선 주파수를 이용하여 리더에서 자동 인식하는 기술이다. RFID 애플리케이션은 RFID 태그의 실시간 정보를 기반으로 하는 업무 프로세스를 의미하는 것으로, RFID 정보를 이용하기 위해서는 기존의 비즈니스 프로세스가 변형되어져야 한다. RFID 미들웨어로부터 발생하는 저수준의 RFID 이벤트를 다양한 정보 서버들을 참조하여 고수준의 이벤트로 변환시키기 위한 일련의 활동(activity)들을 대부분의 RFID 애플리케이션에서 공통으로 수행하기 때문에 이러한 활동들을 재사용 될 수 있는 핵심자산으로 만들어놓을 필요가 있다. 본 논문에서는 다양한 유형의 RFID 애플리케이션에 재사용될 수 있는 RFID 관련 활동들을 RFID 제네릭 (generic) 활동으로 구분하고 이를 이용하여 2-레벨의 비즈니스 프로세스 패밀리 모델 (Business Process Family Model: BPFM)을 구축하는 방법을 제시한다. 상위 레벨의 RFID 제네릭 액티비티들은 두번에 걸쳐 가변치가 결정될 수 있는 가변점을 가지게 된다. 하위 레벨을 구성하는 도메인 활동들은 그 자체의 가변속성 뿐만 아니라 활동의 흐름에서 나타나는 다양한 형태의 가변요소들을 표현하게 된다. 이러한 2-레벨 BPFM을 이용함으로써 RFID 시스템 도입 시 처리해야 하는 활동들의 개발 양을 현저히 줄일 수 있다.

• Smart Media Journal
• /
• v.11 no.2
• /
• pp.9-17
• /
• 2022

The purpose of this study is to implement a predictive model to support decision-making to replace a starting pitcher before a crisis situation in a baseball game. To this end, using the Major League Statcast data provided by Baseball Savant, we implement a predictive model that preemptively replaces starting pitchers before a crisis situation. To this end, first, the crisis situation that the starting pitcher faces in the game was derived through data exploration. Second, if the starting pitcher was replaced before the end of the inning, learning was carried out by composing a label with a replacement in the previous inning. As a result of comparing the trained models, the model based on the ensemble method showed the highest predictive performance with an F1-Score of 65%. The practical significance of this study is that the proposed model can contribute to increasing the team's winning probability by replacing the starting pitcher before a crisis situation, and the coach will be able to receive data-based strategic decision-making support during the game.