• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.175-180
• /
• 2019

관계 추출은 문서에서 존재하는 트리플(주어, 관계어, 목적어)형식에 해당하는 단어를 추출하는 작업을 뜻한다. 본 논문에서는 멀티헤드 셀프 어텐션을 이용하여 트리플 중 주어나 목적어를 찾는 구조를 제안한다. 한국어 위키피디아와 DBpedia의 관계어를 단어 임베딩을 통해 벡터를 생성하고 입력한다. 초록과 관계어의 어텐션 이후 멀티 헤드 셀프 어텐선 구조를 통해 초록 중 관계어와 관련 있는 단어들의 가중치가 높아 진다. 멀티헤드 셀프 어텐션 과정을 반복하여 주요 단어들의 가중치가 계속해서 높아진다. 이를 입력으로 하여 정답이 될 단어의 시작과 끝을 선택한다. 제안 방법으로 직접 구축한 한국어 관계 추출 데이터셋을 대상으로 F1 0.7981의 성능을 보였다. 제안 방법은 관계어와 같이 단순한 정보만을 이용하고도 초록에서 적절한 정답 단어를 추출할 수 있음을 확인하였다. 관계어의 범위를 확장함으로서 나아가 육하원칙(5W1H)과 같은 이벤트의 추출에도 활용할 수 있을 것이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.43-44
• /
• 2023

대화형 독해 튜터링 시스템을 위한 학생주도 대화 데이터셋 생성 및 확장에 ChatGPT 의 활용 가능성을 평가하였다. 단순히 수동으로만 구축한 기존의 데이터셋과 ChatGPT 에 의해 반자동으로 확장된 데이터셋을 비교한 결과, 구축량, 소요 시간, 비용 및 반복 작업 측면에서 ChatGPT 가 가진 유용성을 알 수 있었다. 그러나, 유형별 배분의 편중과, 부적절한 데이터 생성 등의 한계도 나타났다. Chat GPT 의 빠른 발전이 예상됨에 따라 대화형 튜터링 분야에 ChatGPT 에 의한 반자동 데이터 확장 방법이 널리 활용될 것으로 기대된다.

• Journal of Bio-Environment Control
• /
• v.3 no.1
• /
• pp.42-51
• /
• 1994

Visual features of a mushromm(Lentinus Edodes L.) are critical in sorting and grading as most agricultural products are. Because of its complex and various visual features, grading and sorting of mushrooms have been done manually by the human expert. Though actions involved in human grading look simple, it decision making underneath the simple action comes from the result of the complex neural processing of visual image. Recently, an artificial neural network has drawn a great attention because of its functional capability as a partial substitute of the human brain. Since most agricultural products are not uniquely defined in its physical properties and do not have a well defined job structure, the neuro -net based computer visual information processing is the promising approach toward the automation in the agricultural field. In this paper, first, the neuro - net based classification of simple geometric primitives were done and the generalization property of the network was tested for degraded primitives. And then the neuro-net based grading system was developed for a mushroom. A computer vision system was utilized for extracting and quantifying the qualitative visual features of sampled mushrooms. The extracted visual features of sampled mushrooms and their corresponding grades were used as input/output pairs for training the neural network. The grading performance of the trained network for the mushrooms graded previously by the expert were also presented.

• Korean Journal of Construction Engineering and Management
• /
• v.7 no.6
• /
• pp.79-89
• /
• 2006

Current removing process is labor intensive and time consuming, employing a conventional grinding type manual machine. From a social and economic point of views, these kinds of manual tasks bring about social inconvenience and economic loss including traffic jam and high labor costs. The objective of the study was to develop and evaluate a road stripe removal equipment using water jet technology. The following shows the results. First, It was analyzed that the obtainable productivity from the equipment developed in the study is 280% compared to the current equipment. In this study, it was also calculated the Benefit/Cost Ratio and the result showed that the ratio is 3.28, so it is expected that the equipment can produce great benefits for the relevant companies. Second, an analysis was also conducted on the traffic congestion cost, and the equipment could save about \2,550 million per day compared to the conventional equipment. Therefore, it is analyzed that the economic viability of the equipment is sufficient.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.34 no.6
• /
• pp.403-408
• /
• 2021

Currently, in the design process of civil structures such as bridges, it is common to make final products by repeating the process of redesigning, if the initial design is found to not meet the standards after a structural review. This iterative process extends the design time, and causes inefficient consumption of engineering manpower, which should be put into higher-level design, on simple repetitive mechanical work. This problem can be resolved by automating the design process, but the external analysis program used in the design process has been the biggest obstacle to such automation. In this study, we constructed an AI-based automation system for the bridge design process, including an interface that could control both a reinforcement learning algorithm, and an external analysis program, to replace the repetitive tasks in the current design process. The prototype of the system built in this study was developed for a 2-span RC Rahmen bridge, which is one of the simplest bridge systems. In the future, it is expected that the developed interface system can be utilized as a basic technology for linking the latest AI with other types of bridge designs.

• Cartoon and Animation Studies
• /
• s.44
• /
• pp.33-56
• /
• 2016

As the rapid extension of the game and film industry, studies on developing natural movements on a 3D characters are increasing. Rigging a character with joints is essential to create realistic movements on a 3D character. The rapid development of the CG industry, rigging technologies and workflow is becoming more sophisticated. Despite the progress and the growth of rigging operations, has shown the limitations of such repetitive tasks. For this study, analyzes the issues and inefficiency of the old method of skin weights and propose a better approach. First, need to understand the general process of an animation pipeline and learn the technology term of skin weights. Second, comparing the traditional ways of skinning a character and applying other deformers to work properly. Third, testing out new ways of weighting a character by applying deformers such as lattice and finally converting lattice weights back to skin weights. Forth, verifying effectiveness of the new method of skin weights by comparing with the traditional skin weighting process. The study shows that the new method of skin weights, reduced working hours and a better final weighting outcome. Expecting this study to enhance the method of skin weights and able to utilize this new skinning technology.

• Proceedings of the Korean Society of Computational and Applied Mathematics Conference
• /
• 2003.09a
• /
• pp.11-11
• /
• 2003

오늘날 단일 슈퍼컴퓨터로는 처리가 불가능한 거대한 문제들의 해법이 시도되고 있는데, 이들은 지리적으로 분산된 슈퍼컴퓨터, 데이터베이스, 과학장비 및 디스플레이 장치 등을 초고속 통신망으로 연결한 GRID 환경에서 효과적으로 실행시킬 수 있다. GRID는 1990년대 중반 과학 및 공학용 분산 컴퓨팅의 연구 과정에서 등장한 것으로, 점차 응용분야가 넓어지고 있다. 그러나 GRID 같은 분산 환경은 기존의 단일 병렬 시스템과는 많은 점에서 다르며 이전의 기술들을 그대로 적용하기에는 무리가 있다. 기존 병렬 시스템에서는 주로 동기 알고리즘(synchronous algorithm)이 사용되는데, 직렬 연산과 같은 결과를 얻기 위해 동기화(synchronization)가 필요하며, 부하 균형이 필수적이다. 그러나 부하 균형은 이질 클러스터(heterogeneous cluster)처럼 프로세서들의 성능이 서로 다르거나, 지리적으로 분산된 계산자원을 사용하는 GRID 환경에서는 이기종의 문제뿐 아니라 네트워크를 통한 메시지의 전송 지연 등으로 유휴시간이 길어질 수밖에 없다. 이처럼 동기화의 필요성에 의한 연산의 지연을 해결하는 하나의 방안으로 비동기 반복법(asynchronous iteration)이 나왔으며, 지금도 활발히 연구되고 있다. 이는 알고리즘의 동기점을 가능한 한 제거함으로써 빠른 프로세서의 유휴 시간을 줄이는 것이 목적이다. 즉 비동기 알고리즘에서는, 각 프로세서는 다른 프로세서로부터 갱신된 데이터가 올 때까지 기다리지 않고 계속 다음 작업을 수행해 나간다. 따라서 동시에 갱신된 데이터를 교환한 후 다음 단계로 진행하는 동기 알고리즘에 비해, 미처 갱신되지 않은 데이터를 사용하는 경우가 많으므로 전체적으로는 연산량 대비의 수렴 속도는 느릴 수 있다 그러나 각 프로세서는 거의 유휴 시간이 없이 연산을 수행하므로 wall clock time은 동기 알고리즘보다 적게 걸리며, 때로는 50%까지 빠른 결과도 보고되고 있다 그러나 현재까지의 연구는 모두 어떤 수렴조건을 만족하는 선형 시스템의 해법에 국한되어 있으며 비교적 구현하기 쉬운 공유 메모리 시스템에서의 연구만 보고되어 있다. 본 연구에서는 행렬의 주요 고유쌍을 구하는 데 있어 비동기 반복법의 적용 가능성을 타진하기 위해 우선 이론적으로 단순한 멱승법을 사용하여 실험하였고 그 결과 순수한 비동기 반복법은 수렴하기 어렵다는 결론을 얻었다 그리하여 동기 알고리즘에 비동기적 요소를 추가한 혼합 병렬 알고리즘을 제안하고, MPI(Message Passing Interface)를 사용하여 수원대학교의 Hydra cluster에서 구현하였다. 그 결과 특정 노드의 성능이 다른 것에 비해 현저하게 떨어질 때 전체적인 알고리즘의 수렴 속도가 떨어지는 것을 상당히 완화할 수 있음이 밝혀졌다.

• The Journal of the Korea Contents Association
• /
• v.10 no.10
• /
• pp.172-179
• /
• 2010

The study on works motivated from interest in the nature of matters and inherent visual-perceptual structure in them aims at expressing formative continuity of the connections of three dimensions of simple geometrical shapes such as circles and lines, which are characteristics of shape of molecules. With such a purpose, this study examined the geometrical shapes in modern arts and structural connection and symbolism of molecule structure, and based on such considerations, it expressed successive formative beauty which comes from repetitive connection between units by creating stereogram of simple geometrical shapes of molecule structure. The types of works include a method of connecting the units of molecule models and molecules seen in electron microscope with lines as a parameter and connecting units directly, which are used to express body accessory and metallic sculptures. Consequently, it attempted formation occurring spatial composition of continuity of division and duplication through direct connection between units and circular continuity coming from connection of simple geometrical shapes of molecule images such as spheres and curves transformed into stereogram.

• Journal of Practical Engineering Education
• /
• v.14 no.3
• /
• pp.543-553
• /
• 2022

A collaborative robot(Cobot) is one of the production systems presented in the 4th industrial revolution and are systems that can maximize efficiency by combining the exquisite hand skills of workers and the ability of simple repetitive tasks of robots. Also, research on the development of an efficient interface method between the worker and the robot is continuously progressing along with the solution to the safety problem arising from the sharing of the workspace. In this study, a method for controlling the robot by recognizing the worker's hand signal was presented to enhance the convenience and concentration of the worker, and the safety of the worker was secured by introducing the concept of a safety zone. Various technologies such as robot control, PLC, image processing, machine learning, and ROS were used to implement this. In addition, the roles and interface methods of the proposed technologies were defined and presented for using educational media. Students can build and adjust the educational media system by linking the introduced various technologies. Therefore, there is an excellent advantage in recognizing the necessity of the technology required in the field and inducing in-depth learning about it. In addition, presenting a problem and then seeking a way to solve it on their own can lead to self-directed learning. Through this, students can learn key technologies of the 4th industrial revolution and improve their ability to solve various problems.

• Korean Journal of Construction Engineering and Management
• /
• v.13 no.1
• /
• pp.53-66
• /
• 2012

Gangform compared with conventional wood and steel form is systemized form which has a capability to install and dismantle form at a time without repeating to assembly and disassembly small members. Gangform compared with ACS has some problems such as 1) Tower Crane load increase, 2) increase in safety incidence during climbing Gangform, 3) decrease in detachment operation productivity, 4) stop work due to climate impacts, and 5) decrease in labor cost, productivity, quality. A conceptual design model of automated Gangform climbing system is suggested for apartment housing construction to show its technical/economic feasibility and workers' safety while increasing operation productivity and concrete quality.