• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권1호
• /
• pp.81-98
• /
• 2013

Since Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA) were introduced, many revised or extended topic models have appeared. Due to the intractable likelihood of these models, training any topic model requires to use some approximation algorithm such as variational approximation, Laplace approximation, or Markov chain Monte Carlo (MCMC). Although these approximation algorithms perform well, training a topic model is still computationally expensive given the large amount of data it requires. In this paper, we propose a new method, called non-simultaneous sampling deactivation, for efficient approximation of parameters in a topic model. While each random variable is normally sampled or obtained by a single predefined burn-in period in the traditional approximation algorithms, our new method is based on the observation that the random variable nodes in one topic model have all different periods of convergence. During the iterative approximation process, the proposed method allows each random variable node to be terminated or deactivated when it is converged. Therefore, compared to the traditional approximation ways in which usually every node is deactivated concurrently, the proposed method achieves the inference efficiency in terms of time and memory. We do not propose a new approximation algorithm, but a new process applicable to the existing approximation algorithms. Through experiments, we show the time and memory efficiency of the method, and discuss about the tradeoff between the efficiency of the approximation process and the parameter consistency.

• 한국음향학회지
• /
• 제11권3호
• /
• pp.40-52
• /
• 1992

본논문에서는 불특정화자 음성인식의 문제점이 되는 개인차에 의한 변동을 흡수하기 위하여 사상멤버쉽함수에 의한 화자적응 단어인식 방법을 제안하였다. 이방법의 학습과정에서는 미지화자의 표준화자의 스펙트럼패턴 사이에서 작성된 사상코드북에 퍼지이론을 도입하여 사상멤버쉽함수를 작성하였으며, 인식과정에서는 미지화자의 음성패턴을 사상멤버쉽함수에 의해 표준화자의 음성패턴에 적응된 패턴으로 재구성하고 뉴럴-퍼지패턴매칭에 의해 단어를 인식하였다. 본 방법의 타당성을 평가하기 위하여, 28개의 DDD 지역명을 대상으로 실험한 결과, 종래의 사상코드북에 의한 벡터양자화 화자적응방법에서는 64.9[%], 퍼지벡터양자화 화자적응방법에서는 76.1[%]의 인식율을 얻었으나, 사상멤버쉽함수에 의한 화자적응방법에서는 95.4[%]의 향상된 인식율을 얻으므로써 인식성능의 우수함을 확인하였다. 또한 사상멤버쉽함수의 작성과정에서는 반복된 학습과정이 불피요하며, 기억용량과 계산량도 사상코드북에 의한 화자적응방법보다 각각 1/30, 1/500배 정도였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권8호
• /
• pp.2881-2894
• /
• 2014

In complicated environment, context information plays an important role in image segmentation/labeling. The recently proposed auto-context algorithm is one of the effective context-based methods. However, the standard auto-context approach samples the context locations utilizing a fixed radius sequence, which is sensitive to large scale-change of objects. In this paper, we present a scale invariant auto-context (SIAC) algorithm which is an improved version of the auto-context algorithm. In order to achieve scale-invariance, we try to approximate the optimal scale for the image in an iterative way and adopt the corresponding optimal radius sequence for context location sampling, both in training and testing. In each iteration of the proposed SIAC algorithm, we use the current classification map to estimate the image scale, and the corresponding radius sequence is then used for choosing context locations. The algorithm iteratively updates the classification maps, as well as the image scales, until convergence. We demonstrate the SIAC algorithm on several image segmentation/labeling tasks. The results demonstrate improvement over the standard auto-context algorithm when large scale-change of objects exists.

• ETRI Journal
• /
• 제40권3호
• /
• pp.309-317
• /
• 2018

Single carrier-frequency-division multiple access (SC-FDMA) has been adopted as the uplink transmission standard in fourth-generation cellular networks to facilitate power efficiency transmission in mobile stations. Because multiuser multiple-input multiple-output (MU-MIMO) is a promising technology employed to fully exploit the channel capacity in mobile radio networks, this study investigates the uplink transmission of MU-MIMO SC-FDMA systems with orthogonal space-frequency block codes (SFBCs). It is preferable to minimize the length of the cyclic prefix (CP). In this study, the chained turbo equalization technique with chained turbo estimation is employed in the designed receiver. Chained turbo estimation employs a short training sequence to improve the spectrum efficiency without compromising the estimation accuracy. In this paper, we propose a novel and spectrally efficient iterative joint-channel estimation, multiuser detection, and turbo equalization for an MU-MIMO SC-FDMA system without CP-insertion and with short TR. Some simulation examples are presented for the uplink scenario to demonstrate the effectiveness of the proposed scheme.

• Journal of Electrical Engineering and Technology
• /
• 제11권1호
• /
• pp.1-8
• /
• 2016

It is well known that the behavior of PV curves is similar to a quadratic function. This is used in some papers to approximate PV curves and calculate the maximum-loading point by minimum number of power flow runs. This paper also based on quadratic approximation of the PV curves is aimed at completing previous works so that the computational efforts are reduced and the accuracy is maintained. To do this, an iterative method based on a quadratic function with two constant coefficients, instead of the three ones, is used. This simplifies the calculation of the quadratic function. In each iteration, to prevent the calculations from diverging, the equations are solved on the assumption that voltage magnitude at a selected load bus is known and the loading factor is unknown instead. The voltage magnitude except in the first iteration is selected equal to the one at the nose point of the latest approximated PV curve. A method is presented to put the mentioned voltage in the first iteration as close as possible to the collapse point voltage. This reduces the number of iterations needed to determine the maximum-loading point. This method is tested on four IEEE test systems.

• 한국음향학회지
• /
• 제24권5호
• /
• pp.262-270
• /
• 2005

본 논문에서는 음성 신호를 시간축으로 분할하는 새로운 기법으로, 분할 시 왜곡과 엔트로피가 함께 고려된 기법이 제안되었다 시간축 분할에 필요한 보간 함수와 타겟 특징 벡터는 동적 프로그래밍 기법을 이용하여 왜곡과 엔트로피가 동시에 최소화되도록 얻어진다. 보간 함수는 학습 데이터를 이용하여 구성되도록 하였으며, 분할과 추정의 반복적인 수행에 의해 왜곡과 엔트로피가 지역적으로 최소화 되는 지점에서 설계되도록 하였다. 모의 실험에서 제안된 시간축 분할 기법은 현존 음성 부호화 기법에 널리 사용되고 있는 분할 벡터 양자화 기법과 비교하여, 왜곡-비트율 특성 관점에서 보다 우수한 성능을 나타내었으며, 주관적인 청취 테스트 결과, 음질적인 면에서도 기존의 벡터 양자화 기법에 비해 우수한 방법임을 알 수 있었다.

• 지능정보연구
• /
• 제18권2호
• /
• pp.131-141
• /
• 2012

논문에서는 협업필터링(collaborative filtering : CF) 기반한 추천시스템의 정확도를 높일 수 있는 방법을 제안하고 그 효과를 분석한다. 일반적인 CF기반 추천시스템에서는 시스템 세팅(참조집단 크기, 유의도 수준 등)을 한 가지 정해서 모든 경우에 대해서 동일하게 적용한다. 본 논문에서는 개별 사용자의 특성에 따라 이러한 세팅을 최적화 해서 개별적으로 적용하는 방법을 개발하였다. 이런 개인화된 세팅의 효과를 측정하기 위해서 Netflix의 자료를 사용해서 일반적인 추천시스템과 추천 정확도를 비교하였다. 분석 결과, 동일한 세팅을 적용하는 일반적인 추천시스템에 비해서 개인화된 세팅을 적용한 경우 정확도가 월등히 향상됨을 확인하였다. 이 결과의 시사점과 함께 미래 연구의 방향에 대해서도 논의한다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 하계학술대회 논문집 D
• /
• pp.2726-2728
• /
• 2001

This paper describes an optimal synthesis method of binary neural network(BNN) for an approximation problem of a circular region using a newly proposed learning algorithm[7] Our object is to minimize the number of connections and neurons in hidden layer by using a Newly Expanded and Truncated Learning Algorithm(NETLA) for the multilayer BNN. The synthesis method in the NETLA is based on the extension principle of Expanded and Truncated Learning(ETL) and is based on Expanded Sum of Product (ESP) as one of the boolean expression techniques. And it has an ability to optimize the given BNN in the binary space without any iterative training as the conventional Error Back Propagation(EBP) algorithm[6] If all the true and false patterns are only given, the connection weights and the threshold values can be immediately determined by an optimal synthesis method of the NETLA without any tedious learning. Futhermore, the number of the required neurons in hidden layer can be reduced and the fast learning of BNN can be realized. The superiority of this NETLA to other algorithms was proved by the approximation problem of one circular region.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2011년도 춘계학술발표대회
• /
• pp.447-450
• /
• 2011

This paper shows the application of P-N Learning [4] method in the soccer ball detection and improvement for increasing the speed of processing. In the P-N learning, the learning process is guided by positive (P) and negative (N) constraints which restrict the labeling of the unlabeled data, identify examples that have been classified in contradiction with structural constraints and augment the training set with the corrected samples in an iterative process. But for the long-view in the soccer game, P-N learning will produce so many ferns that more time is spent than other methods. We propose that color histogram of each frame is constructed to delete the unnecessary details in order to decreasing the number of feature points. We use the mask to eliminate the gallery region and Line Hough Transform to remove the line and adjust the P-N learning's parameters to optimize accurate and speed.

• Child Health Nursing Research
• /
• 제30권2호
• /
• pp.142-153
• /
• 2024

Purpose: This study aimed to describe the process of developing a validated pediatric nursing simulation scenario template using the real-time Delphi method. Methods: A panel of 13 pediatric nursing experts participated in a real-time Delphi survey conducted over two rounds. Initially, 83 items were included in the questionnaire focusing on the structure and content of the simulation scenario template. Data analysis involved calculating the content validity ratio (CVR) and the coefficient of variation to assess item validity and stability. Results: Through iterative rounds of the Delphi survey, a consensus was reached among the experts, resulting in the development of a pediatric nursing simulation scenario template comprising 41 items across nine parts. The CVR values ranged from 0.85 to 1.0, indicating a high consensus among experts regarding the inclusion of all items in the template. Conclusion: This study presents a novel approach for developing a pediatric nursing simulation scenario template using real-time Delphi methods. The real-time Delphi method facilitated the development of a comprehensive and scientifically grounded pediatric nursing simulation scenario template. Our template aligns with the International Nursing Association for Clinical Simulation and Learning standards, and provides valuable guidance for educators in designing effective simulation scenarios, contributing to enhanced learning outcomes and better preparation for pediatric clinical practice. However, consideration of cultural and contextual adaptations is necessary, and further research should explore alternative consensus criteria.