• Electronics and Telecommunications Trends
• /
• v.36 no.2
• /
• pp.65-74
• /
• 2021

In this paper, we discussed the necessity and importance of introducing feature stores to establish a collaborative environment between data engineering work and data science work. We examined the technology trends of feature stores by analyzing the status of some major feature stores. Moreover, by introducing a feature store, we can reduce the cost of performing artificial intelligence (AI) projects and improve the performance and reliability of AI models and the convenience of model operation. The future task is to establish technical requirements for establishing a collaborative environment between data engineering work and data science work and develop a solution for providing a collaborative environment based on this.

• Food Science and Industry
• /
• v.54 no.3
• /
• pp.160-170
• /
• 2021

A recent advance in communication technologies accelerates the spread of food safety issues once presented by the news media. To respond to those safety issues and take steps in a timely manner, automatically detecting related information from the news data matters. This work presents an AI-based system that detects risk information within a food-related news article. Experts in food safety areas participated in labeling risk information from the food-related news articles; we acquired 43,527 articles in which food names and risk information are marked as labels. Based on the news document, our system automatically detects food names and risk information by analyzing similarities between words within a text by leveraging learned word embedding vectors. Our AI-based system shows higher detection accuracy scores over a non-AI rule-based system: achieving an absolute gain of +32.94% in F1 for the food name category and +41.53% for the risk information category.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.3
• /
• pp.302-308
• /
• 2021

Developers generally use test standards suggested by military standards such as MIL-STD-810G when performing vibration tests in the materiel development. However, according to MIL-STD-810G, it is recommended to test by tailoring the test standard suitable for the developed materiel, and it is specified to apply the suggested test standard only when there is difficulty in tailoring. In addition, the test standards presented by MIL-STD-810G are standards created under operating conditions different from the actual operating environment of each developed materiel, so the test according to this standard may be excessive or understated. Therefore, the developer must create an appropriate vibration test standard for the developed materiel as similar to the operating conditions as possible. In this paper, the procedure for creating the functional test standard and durability test standard suitable for the operating environment of the equipment to be mounted on the propeller aircraft under development is described, and the created standard is introduced.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.14 no.4
• /
• pp.111-121
• /
• 2004

With rapid growth of the Internet, network intrusions have greatly increased and damage of attacks has become more serious. Recently some kinds of Internet attacks cause significant damage to overall network performance. Current Intrusion Detection Systems are not capable of performing the real-time detection on the backbone network In this paper, we propose the broadband network intrusion detection system using the exponential smoothing method. We made an experiment with real backbone traffic data for 8 days. The results show that our proposed system detects big jumps of traffic volume well.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.1
• /
• pp.145-150
• /
• 2021

This thesis attempted to examine the mechanization of humans, humanization of machines, and coexistence through dance works. The dance works were reviewed by partial excerpts from Oscar Schlemer's <3 Chord Ballet>, Felindrome Dance Company's , and . Also, I looked at the dance work , which has an inherent form of coexistence. Through the above work, robot-like science and technology and fusion. It was found that various dance performances that coexist in complex forms provide continuous creativity to humans, and various forms of sensibility and creative movements based on data make it possible to produce rich performances for humans. This researcher expects numerous works that accept and reflect the changes of the times through the embodied interaction of dance performances with science and technology.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.231-231
• /
• 2023

토양침식은 지표의 토양이 바람이나 물에 의해 분리되어 이송되는 자연현상이다. 우리나라에서는 주로 물에 의한 토양침식이 발생하며 특히 단기간 집중적으로 내리는 강우에 의해 토양침식이 일어난다. 토양의 침식현상은 농경지 유실, 하공구조물에서의 퇴적토 발생, 수질 오염등 다양한 문제를 일으키며 기후변화로 인한 집중강우의 빈도 및 강도 증가는 토양침식에 의한 피해를 증가 시키고 있다. 이러한 문제를 파악하기 위해 경험적 방법에 의해 개발된 범용토양유실공식인 USLE 모형이 널리 사용되고 있으나 연간 토양침식량을 산정하기 위해 개발된 USLE모형은 강우기간이 짧고 강우강도가 높은 집중호우와 같은 단기 강우사상을 모의할 수 없고 모든 지역을 표현하는 데 한계가 있다. 이에 따라 단기 강우사상을 고려할 수 있는 물리기반 침식모형인 SSEM모형을 활용하였다. SSEM모형은 운동파 방정식의 수치해석과 물리적 기반 접근방식을 통해 토양침식과정을 계산하여 집중호우로 인해 발생하는 토양침식을 보다 정확하게 추정할 수 있다. 이러한 모형의 적용성을 확인하기 위해 우리나라의 의암댐유역 선정하였으며, 지형 및 강우 그리고 댐자료 등 기초자료 수집과 수집된 데이터는 연구 대상에 대한 토양침식량 산정 및 매개변수 추정과 보정하는 데 사용되었다. 이 결과 다른 토지이용에 비해 농경지와 나지에서 많은 침식이 일어나며 도심지에서의 퇴적이 발생하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.472-475
• /
• 2020

본 논문에서는 청각장애인에게 자막방송을 제공하기 위하여 오디오 콘텐츠에 등장하는 음향 사건을 분류하는 기법을 제안한다. 제안된 기법은 복수의 잔차 신경망(ResNet)을 연결하는 연쇄잔차(concatenated residual) 신경망 구조를 갖는다. 신경망의 입력 특징을 위해 음성의 멜-주파수 켑스트럼 벡터를 다수의 프레임으로 결합하여 형성한 2 차원 이미지와 전체 프레임에 대한 멜-주파수 켑스트럼 벡터들로부터 얻은 1 차원의 통계 특징벡터를 얻는다. 각각의 입력은 2 차원 잔차 신경망과 1 차원 잔차 신경망으로 모델링되고, 두 개의 잔차 신경망을 연쇄연결(concatenation)하는 구조를 가진 연쇄잔차 신경망으로 구성된다. 성능평가를 위해 수집된 데이터셋으로부터 6-fold 교차검증을 통해 평가한 결과, 85.48%의 분류 정확도를 얻을 수 있었다.

• Annual Conference of KIPS
• /
• 2008.11a
• /
• pp.966-969
• /
• 2008

사용자 프로그램의 I/O 패턴을 분석하거나 파일 시스템의 워크로드를 보다 정확하게 분석하기 위해서 실제 가동중인 파일 시스템의 동적 I/O 로그를 확보하기 위한 연구들이 많이 진행되어 왔다. 그러나 대량의 I/O 트렌젝션(transcation)이 처리되는 파일 시스템에서 동적 I/O 로그를 확보하는 일은 시스템의 부하와 막대한 데이터량 때문에 한계가 많다. 특히 다수의 이용자가 사용하는 대용량 분산/병렬 파일 시스템에서의 I/O Tracing은 로컬 파일 시스템에서 I/O Tracing에 비해 더욱 복잡하고 오버헤드가 크다. 본 논문에서는 기존의 파일 시스템 로깅 방법들을 알아보고, 클러스터 시스템에서 널리 이용되고 있는 분산 파일 시스템인 PVFS(Parallel Virtual File System)에서 동적 I/O 연산들의 로그를 생성할 수 있는 로깅 시스템을 제안하고 설계하였다.

• KIISE Transactions on Computing Practices
• /
• v.21 no.12
• /
• pp.745-750
• /
• 2015

LiDAR(Light Detection and Ranging) technology provides realistic 3-dimension image information, and it has been widely utilized in various fields. However, the utilization of this technology in the military domain requires prompt responses to dynamically changing tactical environment and is therefore limited since LiDAR technology requires complex processing in order for extensive amounts of LiDAR data to be utilized. In this paper, we introduce an Unmanned Aircraft Platform Based Real-time LiDAR Data Processing Architecture that can provide real-time detection information by parallel processing and off-loading between the UAV processing and high-performance data processing areas. We also conducted experiments to verify the feasibility of our proposed architecture. Processing with ARM cluster similar to the UAV platform processing area results in similar or better performance when compared to the current method. We determined that our proposed architecture can be utilized in the military domain for tactical and combat purposes such as unmanned monitoring system.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.785-794
• /
• 2019

Recently Deep Learning technology, one of the fourth industrial revolution technologies, is used to identify the hidden meaning of network data that is difficult to detect in the security arena and to predict attacks. Property and quality analysis of data sources are required before selecting the deep learning algorithm to be used for intrusion detection. This is because it affects the detection method depending on the contamination of the data used for learning. Therefore, the characteristics of the data should be identified and the characteristics selected. In this paper, the characteristics of malware were analyzed using network data set and the effect of each feature on performance was analyzed when the deep learning model was applied. The traffic classification experiment was conducted on the comparison of characteristics according to network characteristics and 96.52% accuracy was classified based on the selected characteristics.