• The Journal of the Acoustical Society of Korea
• /
• v.17 no.4
• /
• pp.3-13
• /
• 1998

본 논문은 음성 다이어링 시스템을 구현하기 위한 한국어 단독 숫자음 및 연속 숫 자음 인식에 관한 것이다. 단독 숫자음의 인식은 미지의 입력 음성을 재귀 신경망을 이용하 여 모델링된 각 모델에 인가하고, 신경 회로망의 출력 노드의 상태열을 검사하여 적절한 상 태 전이를 하며 최고의 확률값을 출력하는 모델을 인식된 결과로 출력한다. 연속 숫자음의 인식은 미지의 연속 숫자음을 재귀 신경 회로망을 이용한 연속 숫자음 모델에 입력하고, 신 경 회로망의 출력에 대하여 적절한 상태 전이에 대한 검사와 레벨 빌딩(Level Building)을 수행하여 최소의 오차를 가지는 모델열을 인식된 결과로 출력한다. 재귀 신경 회로망을 이 용하여 음절 모델을 만드는 과정에서 재귀 노드는 예상치가 주어지지 않으므로 신경 회로망 의 학습에서 제외되어 현저한 학습 속도의 저하를 가져온다. 따라서 본 논문에서는 재귀 신 경 회로망의 학습 속도를 향상시키기 위한 2가지 방법을 제안 한다. 첫 번째는 재귀 신경 회로망의 재귀 노드의 예상치를 실험적으로 주어줌으로써 학습 속도의 향상을 도모하였다. 두 번째는 음절 모델의 출력노드의 개수와 음절 모델의 세그먼트 경계를 알고리듬을 이용하 여 자동적으로 조절하였다. 실험결과, 단독어의 경우 음절 '에'에 포함하는 한국어 11개의 숫 자음에 대하여 화자 종속의 경우 97.3%, 화자 독립의 경우 80.5%의 인식률을 얻었으며, 연 속 숫자음의 경우는 21종류의 연속 숫자음에 대하여 화자 종속에서 88.2%, 화자 독립의 경 우 81.3%의 인식률을 얻을 수 있었다.

• 건강소식
• /
• v.11 no.6 s.103
• /
• pp.40-41
• /
• 1987

• Journal of Intelligence and Information Systems
• /
• v.12 no.1
• /
• pp.107-123
• /
• 2006

We propose the Generalized Binary Second-order Recurrent Neural Networks(GBSRNNf) being equivalent to regular grammars and ?how the implementation of lexical analyzer recognizing the regular languages by using it. All the equivalent representations of regular grammars can be implemented in circuits by using GSBRNN, since it has binary-valued components and shows the structural relationship of a regular grammar. For a regular grammar with the number of symbols m, the number of terminals p, the number of nonterminals q, and the length of input string k, the size of the corresponding GBSRNN is $O(m(p+q)^2)$ and its parallel processing time is O(k) and its sequential processing time, $O(k(p+q)^2)$.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.3
• /
• pp.11-25
• /
• 2012

We present the method to infer Context-Free Grammars by applying genetic algorithm to the Binary Third-order Recurrent Neural Networks(BTRNN). BTRNN is a multiple-layered architecture of recurrent neural networks, each of which is corresponding to an input symbol, and is combined with external stack. All parameters of BTRNN are represented as binary numbers and each state transition is performed with any stack operation simultaneously. We apply Genetic Algorithm to BTRNN chromosomes and obtain the optimal BTRNN inferring context-free grammar of positive and negative input patterns. This proposed method infers BTRNN, which includes the number of its states equal to or less than those of existing methods of Discrete Recurrent Neural Networks, with less examples and less learning trials. Also BTRNN is superior to the recent method of chromosomes representing grammars at recognition time complexity because of performing deterministic state transitions and stack operations at parsing process. If the number of non-terminals is p, the number of terminals q, the length of an input string k, and the max number of BTRNN states m, the parallel processing time is O(k) and the sequential processing time is O(km).

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10b
• /
• pp.655-657
• /
• 1998

합병 문제는 정렬된 두 개의 리스트 A=(a1, a2,....,al)와 B=(b1, b2,....,bm)가 주어질 때, 이두 리스트를 합병하여 하나의 정렬된 리스트 C=(c1, c2,....,cn),n=l+m,로 만드는 문제이다. 지금까지는 {{{{ SQRT {n} }}}}$\times${{{{ SQRT {n} }}}} 메쉬 상에서 odd-even 합병을 재귀적으로 수행하는 O({{{{ SQRT {n} }}}})시간 합병 알고리즘이 주로 사용되었는데, 이 재귀적 합병 알고리즘은 그 기본정책은 간단하나, 메쉬 상에 구현 될때 라우팅과 같은 복잡한 연산이 필요하게 된다. 따라서 본 논문에서는 간단한 행, 열 연산 만을 사용하여 메쉬 상에서 쉽게 구현가능한 O({{{{ SQRT {n} }}}})시 간의 최적 분할 합병 알고리즘을 제안한다.

• Journal of Popular Narrative
• /
• v.26 no.4
• /
• pp.145-181
• /
• 2020

This study analyzes type variations of the 'stepmother' and 'sister' in the full-length novels of Park Kyong-Ni and attempted to point out their meanings. The pattern of "negative stepmother" that appeared in classical and new novels also appeared repeatedly in Park Kyong-Ni's full-length novels and this was because a change took place in later full-length novels. Novels analyzed with focus were Jaegwiyeol(1959), Eunha(1960), Kimyakgukeue Ddaldeul(1962), and Nabiwa Unggungkwi(1969). The stepmother that appears in Eunha is a type that appears often in the classic and new novels of Korea. While the stepmother newly gained the role and status of 'mother', she forms a competitive relationship with the daughter of the former wife while still refusing to be a member of the family and she puts the former wife's daughter in critical situations by committing misdeeds. However, the young stepmother in Nabiwa Unggungkwi actually becomes a victim to the malicious and morbid harassment of the former wife's daughter. This stepmother is a good-natured figure who shows a sense of guilt for failing to fulfill her responsibilities of upbringing and education and she eventually dies as a victim to a bomb during the war, leaving her young biological daughter behind. On one hand, the sisters in Jaegwiyeol and Kimyakgukeue Ddaldeul are not strongly bonded but when one is caught in a crisis, the other one claims to be of help. Unlike this, the sisters in Nabiwa Unggungkwi have a bond that cannot be broken. They are half-sisters that bind each other so severely that they hinder each other's growth and they eventually end up disintegrating. Through such analyses, it is shown that issues of human nature are dealt with more acutely by breaking the 'young stepmother' away from convention by placing her in the position of the victim to amplify the conflicting relationship between sisters, unlike in previous pieces. This study was significant in that it looked into how previously repetitive character type changes appeared in full-length novels in conditions that clearly display the writer's determination to leave behind a masterpiece.

• Applied Chemistry for Engineering
• /
• v.28 no.1
• /
• pp.35-41
• /
• 2017

Materials exposed to outside will deteriorate due to various weathering factors such as sunlight, heat, temperature, humidity and so on. Therefore, predicting speed of degradation and life time is a very important issue. This research uses retro-reflective sheets with white and green which are most commonly used colors to conduct the outdoor exposure test and acceleration test of xenon arc in Arizona state in the United States, Chennai in India, Sanary in France and Seosan in Korea to measure the reflective performance of retro-reflection. The accelerated factor was obtained by using regression analysis through reflective values obtained from the acceleration test of xenon arc from Seosan area. Also, by using solar uv program, the accelerated factor of various climate regions were obtained and it was confirmed that the accelerated factor of Senary was 1.04, Arizona 1.82, Chennai 1.92 times higher than that of Seosan.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.1
• /
• pp.11-18
• /
• 2022

De novo drug design is the process of developing new drugs that can interact with biological targets such as protein receptors. Traditional process of de novo drug design consists of drug candidate discovery and drug development, but it requires a long time of more than 10 years to develop a new drug. Deep learning-based methods are being studied to shorten this period and efficiently find chemical compounds for new drug candidates. Many existing deep learning-based drug design models utilize recurrent neural networks to generate a chemical entity represented by SMILES strings, but due to the disadvantages of the recurrent networks, such as slow training speed and poor understanding of complex molecular formula rules, there is room for improvement. To overcome these shortcomings, we propose a deep learning model for SMILES string generation using variational autoencoders with self-attention mechanism. Our proposed model decreased the training time by 1/26 compared to the latest drug design model, as well as generated valid SMILES more effectively.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.10a
• /
• pp.301-305
• /
• 2003

최근에 R은 여러 분야에서 많이 사용되고 있다. 특히 모의실험(simulation)이나 통계학 관련 연구에 많이 사용되고 있다. 모의실험을 하는 경우에는 많은 반복으로 인해 R 프로그램의 수행 속도가 매우 중요하다. 또한 데이터마이닝 분야에서도 R을 많이 사용하고 있다. 우리는 데이터 마이닝에서 데이터의 전처리 과정 중 Fayyad & Irani 방법을 사용하여 연속형 변수를 이산화하는 실험을 하였으며, 이를 위해 R을 사용하였다. 이 프로그램은 재귀 함수를 이용하고 이런 과정에서 빈도표 작성, information계산, 빈도표의 분할, 정지 규칙 등의 여러 함수를 사용하게 되어있다. 우리가 작성한 R 로드를 사용하여 UCI DB의 Iono 자료를 (속성이 35개, 사례수가 약 1000개정도) 이산화 하였을 때 7초 이상의 상당한 시간이 소요된다. 반면에 JAVA로 만들어진 Weka에서 똑같은 Fayyad & Irani 방법을 수행했을 때 위와 같은 큰 자료를 이산화하는 속도가 매우 빨라 수행시간은 거의 무시할 만하였다. 이런 차이점을 보고 R 프로그램의 수행 속도를 늘이는 방법을 찾게 되었다. 이 본 발표에서는 R 코드 중 시간이 많이 소요되는 것들을 몇 가지 선정하고 이들을 더 효율적으로 만들 수 있는 코드를 작성하여 이들 코드의 수행속도를 비교하였다. 또한 몇 가지 명령에 대해서는SAS와도 비교하였다.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.279-280
• /
• 2008

In order to improve the performance and simplify the structure of the conventional detectors in high density magnetic channels, a new equalizer based on bilinear recursive polynomial (BRP) models, which uses the previously estimated sequence, is proposed. The performance is compared with the conventional equalizers and the maximum likelihood sequence detection (MLSD) bound.