• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2853-2873
• /
• 2013

Over the last few years, one of the most common purposes of smart homes is to provide human centric services in the domain of u-healthcare by analyzing inhabitants' daily living. Currently, the major challenges in activity recognition include the reliability of prediction of each classifier as they differ according to smart homes characteristics. Smart homes indicate variation in terms of performed activities, deployed sensors, environment settings, and inhabitants' characteristics. It is not possible that one classifier always performs better than all the other classifiers for every possible situation. This observation has motivated towards combining multiple classifiers to take advantage of their complementary performance for high accuracy. Therefore, in this paper, a method for activity recognition is proposed by optimizing the output of multiple classifiers with Genetic Algorithm (GA). Our proposed method combines the measurement level output of different classifiers for each activity class to make up the ensemble. For the evaluation of the proposed method, experiments are performed on three real datasets from CASAS smart home. The results show that our method systematically outperforms single classifier and traditional multiclass models. The significant improvement is achieved from 0.82 to 0.90 in the F-measures of recognized activities as compare to existing methods.

• Journal of Microbiology and Biotechnology
• /
• 제1권1호
• /
• pp.54-62
• /
• 1991

Cyclodextrin glucanotransferase(CGTase) derived from Bacillus macerans was immobilized by (1) covalent linkage on chitosan and chitin with glutaraldehyde, (2) adsorption on DEAE-cellulose and Amberite IRA 900 after succinylation, and (3) entrapment on alginate and polyacrylamide by cross linking. Adsorption on Amberite IRA 900 and covalent linking on chitosan were identified to be the most suitable immobilization methods considering the yield of activity and stability of immobilized CGTase. The enzymatic properties of immobilized CGTase were investigated and compared with those of the soluble CGTase. Thermal stability of CGTase immobilized on chitosan was increased from 50 to $55^{\circ}C$, and the optimum temperature of CGTase immobilized on Amberite IRA 900 was shifted from 55 to $50^{\circ}C$. The effect of molecular size of soluble starch (substrate) on immobilized CGTase investigated using partially liquefied substrates with different dextrose equivalent(DE). Cyclodextrin(CD) conversion yield augmented according to the increase of DE level for immobilized CGTase on Amberite IRA 900. CD conversion yield of partially cyclized starch with soluble CGTase was higher compared with liquefied one with ${\alpha}-amylase$.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1080-1085
• /
• 2005

A hybrid control scheme is proposed for the stabilization of backward movement along simple paths for a vehicle composed of a truck and six trailers. The hybrid comprises the combination of a linear quadratic regulator (LQR) and a neurocontroller (NC) that is trained by a genetic algorithm (GA). Acting singly, either the NC or the LQR are unable to perform satisfactorily over the entire range of the operation required, but the proposed hybrid is shown to be capable of providing good overall system performance. The evaluation function of the NC in the hybrid design has been modified from the conventional type to incorporate both the squared errors and the running steps errors. The reverse movement of the trailer-truck system can be modeled as an unstable nonlinear system, with the control problem focusing on the steering angle. Achieving good backward movement is difficult because of the restraints of physical angular limitations. Due to these constraints the system is impossible to globally stabilize with standard smooth control techniques, since some initial states necessarily lead to jack-knife locks. This paper demonstrates that a hybrid of neural networks and LQR can be used effectively for the control of nonlinear dynamical systems. Results from simulated trials are reported.

• 한국인터넷방송통신학회논문지
• /
• 제8권5호
• /
• pp.197-203
• /
• 2008

본 연구에서는 차세대 CDMA 셀룰러 네트워크의 성능향상을 위한 멀티 데이터 전송 레이트 조절과 전력제어 문제를 결합한 문제를 제시하고 유전 알고리즘 접근에 기반한 새로운 해법을 제안하였다. 본 연구에서 제안한 알고리즘으로 많은 사용자를 보다 빨리 서비스 할 수 있고 SIR 제한값을 고려하면서 최적의 해를 얻을 수 있었다. 이 연구에서는 CDMA 셀룰러 네트워크의 데이터 전송 레이트를 최대화하는 적합도 문제와 가능한 모바일의 수를 최대화하는 적합도 문제 두 개를 고려하였으나 이 들을 결합한 문제나 다른 것 들에 대한 적합도 함수도 충분히 장착이 가능하다. 시뮬레이션 결과는 접근 방식의 유효성과 타당성을 보여준다.

• 한국지반공학회논문집
• /
• 제21권9호
• /
• pp.25-34
• /
• 2005

본 연구에서는 최적의 인공신경망 구조 설계를 위하여 인공신경망과 유전자 알고리즘이 결합된 신경망구조 설계기법이 제안되었다. 저자들은 신경망 구조설계시 인공지능 적용에 따른 계산적인 복잡함을 줄이며, 신경망에 의한 예측의 정확성을 증가시키기 위하여 인공신경망과 유전자 알고리즘의 특성을 조합하였다. 최적의 신경망 구조를 얻기 위하여 신경망 구조의 설계변수들에 대한 유전자 선별기법을 적용하였다. 제안된 합성 기법의 적용성을 평가하기 위하여 여러 지반공학 물성치들을 추정하는 해석에 적용되었다.

• Structural Engineering and Mechanics
• /
• 제63권2호
• /
• pp.237-249
• /
• 2017

The Self-Organization Feature Map as an unsupervised network is very widely used these days in engineering science. The applied network in this paper is the Self Organization Feature Map with constant weights which includes Kohonen Network. In this research, Reinforced Concrete Shear Wall buildings with different stories and heights are analyzed and a database consisting of measured fundamental periods and characteristics of 78 RC SW buildings is created. The input parameters of these buildings include number of stories, height, length, width, whereas the output parameter is the fundamental period. In addition, using Genetic Algorithm, the structure of the Self-Organization Feature Map algorithm is optimized with respect to the numbers of layers, numbers of nodes in hidden layers, type of transfer function and learning. Evaluation of the SOFM model was performed by comparing the obtained values to the measured values and values calculated by expressions given in building codes. Results show that the Self-Organization Feature Map, which is optimized by using Genetic Algorithm, has a higher capacity, flexibility and accuracy in predicting the fundamental period.

• Asian-Australasian Journal of Animal Sciences
• /
• 제30권12호
• /
• pp.1718-1723
• /
• 2017

Objective: The study explored the dynamics of shearing force and its correlation with chemical compositions and in vitro dry matter digestibility (IVDMD) of stylo. Methods: The shearing force, diameter, linear density, chemical composition, and IVDMD of different height stylo stem were investigated. Linear regression analysis was done to determine the relationships between the shearing force and cut height, diameter, chemical composition, or IVDMD. Results: The results showed that shearing force of stylo stem increased with plant height increasing and the crude protein (CP) content and IVDMD decreased but fiber content increased over time, resulting in decreased forage value. In addition, tall stem had greater shearing force than short stem. Moreover, shearing force is positively correlated with stem diameter, linear density and fiber fraction, but negatively correlated with CP content and IVDMD. Conclusion: Overall, shearing force is an indicator more direct, easier and faster to measure than chemical composition and digestibility for evaluation of forage nutritive value related to animal performance. Therefore, it can be used to evaluate the nutritive value of stylo.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권5호
• /
• pp.455-464
• /
• 2007

지문인식 시스템의 성능을 평가하기 위해서는 대규모 지문 DB를 구축하는 것이 필요하다. 지문을 수집하는 것은 매우 고비용의 작업이기 때문에, 지문 평가용 DB의 구축은 많이 이루어지지 않았고 실제로 소수의 평가용 DB만이 공개되어 있다. 뿐만 아니라 이들 DB는 제한된 환경에서 수집되어 있어 실제 다양한 환경에 대한 지문인식 시스템의 성능을 정확히 평가하기가 어렵다. 본 논문에서는 유전자 알고리즘을 이용하여 소수의 학습 샘플로부터 실제 환경에서 발생하는 다양한 영향을 고려한 지문영상을 자동으로 생성하는 방법을 제안한다. 제안하는 방법을 이용하여 생성된 지문은 실제 환경에서 수집된 지문과 유사한 특성을 가지기 때문에, 실제로 다수의 지문 영상을 수집하지 않고도 대상 환경에서의 성능평가가 가능하다. 실제 지문과의 비교를 통하여 제안하는 방법의 유용성을 검증하였다.

• Clinical and Experimental Pediatrics
• /
• 제58권11호
• /
• pp.407-414
• /
• 2015

Early-onset epileptic encephalopathies are one of the most severe early onset epilepsies that can lead to progressive psychomotor impairment. These syndromes result from identifiable primary causes, such as structural, neurodegenerative, metabolic, or genetic defects, and an increasing number of novel genetic causes continue to be uncovered. A typical diagnostic approach includes documentation of anamnesis, determination of seizure semiology, electroencephalography, and neuroimaging. If primary biochemical investigations exclude precipitating conditions, a trial with the administration of a vitaminic compound (pyridoxine, pyridoxal-5-phosphate, or folinic acid) can then be initiated regardless of presumptive seizure causes. Patients with unclear etiologies should be considered for a further workup, which should include an evaluation for inherited metabolic defects and genetic analyses. Targeted next-generation sequencing panels showed a high diagnostic yield in patients with epileptic encephalopathy. Mutations associated with the emergence of epileptic encephalopathies can be identified in a targeted fashion by sequencing the most likely candidate genes. Next-generation sequencing technologies offer hope to a large number of patients with cryptogenic encephalopathies and will eventually lead to new therapeutic strategies and more favorable long-term outcomes.

• 산업경영시스템학회지
• /
• 제41권2호
• /
• pp.105-116
• /
• 2018

In this study, we consider the assembly line balancing (ALB) problem which is known as an very important decision dealing with the optimal design of assembly lines. We consider ALB problems with soft constraints which are expected to be fulfilled, however they are not necessarily to be satisfied always and they are difficult to be presented in exact quantitative forms. In previous studies, most researches have dealt with hard constraints which should be satisfied at all time in ALB problems. In this study, we modify the mixed integer programming model of the problem introduced in the existing study where the problem was first considered. Based on the modified model, we propose a new algorithm using the genetic algorithm (GA). In the algorithm, new features like, a mixed initial population selection method composed of the random selection method and the elite solutions of the simple ALB problem, a fitness evaluation method based on achievement ratio are applied. In addition, we select the genetic operators and parameters which are appropriate for the soft assignment constraints through the preliminary tests. From the results of the computational experiments, it is shown that the proposed algorithm generated the solutions with the high achievement ratio of the soft constraints.