http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
김현경(Hyungyung Kim),오영섭(Youngsup Oh),송대준(Daejun Song),오승택(Sengtaek Oh),김동학(Donghak Kim),이동열(Dongyul Lee) 한국자동차공학회 2014 한국자동차공학회 부문종합 학술대회 Vol.2014 No.5
A new point of view in vehicle bumper frame is proposed by Contacted 2 Pipes BUMPER STRUCTURE, which is a new development method based on the principle of hot-stamping, structural design, method of TRIZ and evaluations. As a result of the application of new bumper frame, this paper classifies Increased of strength to 1.8GPa, Shapes of Cold Coiled Thin Pipe, optimized structural design, and enhanced performance of RCAR, Towing and Shipping.
Zero-anaphora resolution in Korean based on deep language representation model: BERT
Kim, Youngtae,Ra, Dongyul,Lim, Soojong Electronics and Telecommunications Research Instit 2021 ETRI Journal Vol.43 No.2
It is necessary to achieve high performance in the task of zero anaphora resolution (ZAR) for completely understanding the texts in Korean, Japanese, Chinese, and various other languages. Deep-learning-based models are being employed for building ZAR systems, owing to the success of deep learning in the recent years. However, the objective of building a high-quality ZAR system is far from being achieved even using these models. To enhance the current ZAR techniques, we fine-tuned a pretrained bidirectional encoder representations from transformers (BERT). Notably, BERT is a general language representation model that enables systems to utilize deep bidirectional contextual information in a natural language text. It extensively exploits the attention mechanism based upon the sequence-transduction model Transformer. In our model, classification is simultaneously performed for all the words in the input word sequence to decide whether each word can be an antecedent. We seek end-to-end learning by disallowing any use of hand-crafted or dependency-parsing features. Experimental results show that compared with other models, our approach can significantly improve the performance of ZAR.
이동열(Dongyul Lee),김송정(Songjung Kim),임현준(Hyunjune Yim) 대한기계학회 2005 대한기계학회 춘추학술대회 Vol.2005 No.11
All manufactured parts and tooling have unavoidable variations from their nominal shapes. During assembly, compliant parts are further deformed by relatively rigid assembly tooling. Lack of Knowledge regarding variations and deformations often results in expensive problems. Most current computer-aided design systems are based on ideally sized, ideally located and rigid geometry. This paper proposes a model for the assembly of compliant, non-ideal part. We start by defining tolerance analysis as the process simulation which is the variation of a product or a subassembly when given the tolerance of required parts. Analysis is then done by finite element analysis and using the material properties of the actual parts to be assembled. Using the result, the weld process is estimated.
이동열(Dongyul Lee),이재열(Jaeyeol Lee),김국생(Kuksaeng Kim),임현준(Hyunjune Yim),이규봉(Gyu Bong Lee) 한국자동차공학회 2003 한국자동차공학회 춘 추계 학술대회 논문집 Vol.- No.-
A methodology of analyzing dimensional variations of a subassembly has been developed. considering the effects of weld distortion and weld sequence. As an example, the tailgate of a sports utility vehicle has been considered. A complete thermo-elasto-plastic<br/> finite element analysis has been conducted to find the weld distortion. Using the results.<br/> tolerance analyses have been conducted to find the effects of different weld sequences. Variances of the measured dimensions on the subassembly have been found to be affected by the weld sequence. in spite of the small magnitudes of weld distortion.
이동열(Dongyul Lee),김송정(Songjung Kim),임현준(Hyunjune Yim) 한국자동차공학회 2005 한국자동차공학회 Symposium Vol.- No.-
All manufactured parts and tooling have unavoidable variations from their nominal shapes. During assembly, compliant parts are further deformed by relatively rigid assembly tooling. Lack of Knowledge regarding variations and deformations often results in expensive problems. Most current computer-aided design systems are based on ideally sized, ideally located and rigid geometry. This paper proposes a model for the assembly of compliant, non-ideal part. We start by defining tolerance analysis as the process simulation which is the variation of a product or a subassembly when given the tolerance of required parts. Analysis is then done by finite element analysis and using the material properties of the actual parts to be assembled. Using the result, the weld process is estimated.
초성 우선배열 방식을 이용한 한글 초성 와일드카드 질의 고속 탐색 기법
김영태(Youngtae Kim),강동민(Dongmin Kang),박상우(Sangwoo Park),나동열(Dongyul Ra ) 한국정보과학회 2011 정보과학회 컴퓨팅의 실제 논문지 Vol.17 No.10
자동차의 내비게이터, 휴대폰, PDA 등에서 지역 검색 작업의 경우 기기의 특성상 검색어 타이핑 입력이 쉽지 않다. 따라서 이러한 상황에서 한글 초성을 이용하는 와일드카드 검색이 빈번하다. 게다가 전국 또는 대도시의 경우 검색 대상이 되는 지역 명칭의 수는 방대하여 효율적인 검색 기법이 중요하다. 본 논문은 이러한 초성을 사용하는 와일드카드 질의에 대하여 고속으로 명칭을 검색하여 주는 알고리즘을 소개한다. 이 알고리즘은 검색 속도를 높이기 위하여 “초성우선배열”이라는 새로운 사전 배열 방식을 이용한다. 그 결과 기존의 사전배열 방식을 이용하는 것과 비교할 때 높은 속도 향상을 이룰 수 있었다. 특히 질의 안에 초성 와일드카드 글자의 비중이 높을수록 더욱 속도 향상의 정도가 증가하였다. It is not easy in location search task to input search terms with typing when we use car navigators, cell phones, PDA, etc. because of device characteristics. Therefore, wildcard search with initial consonants in Korean(Hangul Chosung) is frequently used in these circumstances. Moreover, the number of location terms is huge in case of a whole country or a big city and thus efficient retrieval method is important. In this paper, we introduce an algorithm that enables high-speed retrieval for wildcard queries that use initial consonants. This algorithm utilizes a new lexicographic ordering scheme giving first priority to initial consonants to enhance retrieval speed, which resulted in great performance improvement compared with traditional lexicographic ordering. It was especially observed that increasing the number of initial consonant wildcard letters in a query led to faster search.