• Journal of information and communication convergence engineering
• /
• v.3 no.1
• /
• pp.43-48
• /
• 2005

In this paper, the adaptive fuzzy logic controller(AFLC) is proposed, which uses real-coding genetic algorithm showing a good performance on convergence velocity and diversity of population among evolutionary computations. The effectiveness of the proposed AFLC was demonstrated by computer simulation for speed control system of AC servo motor. As a result of simulation for the AC servo motor, it is shown the proposed AFLC has the better performance on overshoot, settling time and rising time than the PI controller which is used when tuning AFLC.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.583-585
• /
• 2004

암의 발병을 조기에 예측하고 진단하는 것은 매우 중요하지만 그 과정이 매우 복잡하고 많은 노력이 필요하다. 암이 발생하는 원인은 매우 다양하지만 근본적으로 단백질을 형성하는 유전자에 변화가 오기 때문으로 생각해 볼 수 있다. 유전자 발현 정보로부터 기계적으로 암을 예측하기 위한 과정은 중요한 유전자의 선택, 모델의 학습, 모델을 이용한 예측과정으로 나뉘어 진다. 본 논문에서는 대장암 여부를 유전자 발현 데이터로부터 예측하기 위한 종분화 진화 신경망을 제안한다. 종분화 진화 신경망은 진화 알고리즘을 사용하여 신경망의 구조를 결정하고 종분화 알고리즘을 사용하여 다양한 개체의 생성을 유도한 후 모델의 앙상블을 통해 보다 높은 성능을 내는 방법이다 실험 결과 제안하는 방법이 대장암 예측 cross validation 테스트에서 96.5%의 높은 성능을 보였다.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.327-338
• /
• 2022

Classification models for human action recognition require robust features and large training sets for good generalization. However, data augmentation methods are employed for imbalanced training sets to achieve higher accuracy. These samples generated using data augmentation only reflect existing samples within the training set, their feature representations are less diverse and hence, contribute to less precise classification. This paper presents new data augmentation and action representation approaches to grow training sets. The proposed approach is based on two fundamental concepts: virtual video generation for augmentation and representation of the action videos through robust features. Virtual videos are generated from the motion history templates of action videos, which are convolved using a convolutional neural network, to generate deep features. Furthermore, by observing an objective function of the genetic algorithm, the spatiotemporal features of different samples are combined, to generate the representations of the virtual videos and then classified through an extreme learning machine classifier on MuHAVi-Uncut, iXMAS, and IAVID-1 datasets.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.78.3-78.3
• /
• 2019

We develop a novel technique that can constrain the evolutionary track of the epoch of reionization (EoR) by applying the convolutional neural network (CNN) to the 21-cm differential brightness temperature. We use 21cmFAST, a fast semi-numerical cosmological 21-cm signal simulator, to produce mock 21-cm map between z=6-13. We design a CNN architecture that predicts the volume-averaged neutral hydrogen fraction from the given 21-cm map. The estimated neutral fraction has a good agreement with its truth value even after smoothing the 21-cm map with somewhat realistic choices of beam size and the frequency bandwidth of the Square Kilometre Array (SKA). Our technique could be further utilized to denoise the 21-cm map or constrain the properties of the radiation sources.

• The Bulletin of The Korean Astronomical Society
• /
• v.45 no.1
• /
• pp.66.2-66.2
• /
• 2020

We propose a deep learning analysis technique with a convolutional neural network (CNN) to predict the evolutionary track of the Epoch of Reionization (EoR) from the 21-cm differential brightness temperature tomography images. We use 21cmFAST, a fast semi-numerical cosmological 21-cm signal simulator, to produce mock 21-cm maps between z = 6 ~ 13. We then apply two observational effects, such as instrumental noise and limit of (spatial and depth) resolution somewhat suitable for realistic choices of the Square Kilometre Array (SKA), into the 21-cm maps. We design our deep learning model with CNN to predict the sliced-averaged neutral hydrogen fraction from the given 21-cm map. The estimated neutral fraction from our CNN model has great agreement with the true value even after coarsely smoothing with broad beam size and frequency bandwidth and heavily covered by noise with narrow beam size and frequency bandwidth. Our results show that the deep learning analyzing method has the potential to reconstruct the EoR history efficiently from the 21-cm tomography surveys in future.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.1
• /
• pp.26-32
• /
• 2018

The protein secondary structures are important information for studying the evolution, structure and function of proteins. Recently, deep learning methods have been actively applied to predict the secondary structure of proteins using only protein sequence information. In these methods, widely used input features are protein profiles transformed from protein sequences. In this paper, to obtain an effective protein profiles, protein profiles were constructed using protein sequence search methods such as PSI-BLAST and HHblits. We adjust the similarity threshold for determining the homologous protein sequence used in constructing the protein profile and the number of iterations of the profile construction using the homologous sequence information. We used the protein profiles as inputs to convolutional neural networks and recurrent neural networks to predict the secondary structures. The protein profile that was created by adding evolutionary information only once was effective.

• Structural Engineering and Mechanics
• /
• v.11 no.5
• /
• pp.485-502
• /
• 2001

Modem soft computing methods, such as neural networks, evolutionary models and fuzzy logic, are mainly inspired by the problem solving strategies the biological systems use in nature. As such, the soft computing methods are fundamentally different from the conventional engineering problem solving methods, which are based on mathematics. In the author's opinion, these fundamental differences are the key to the full understanding of the soft computing methods and in the realization of their full potential in engineering applications. The main theme of this paper is to discuss the fundamental differences between the soft computing methods and the mathematically based conventional methods in engineering problems, and to explore the potential of soft computing methods in new ways of formulating and solving the otherwise intractable engineering problems. Inverse problems are identified as a class of particularly difficult engineering problems, and the special capabilities of the soft computing methods in inverse problems are discussed. Soft computing methods are especially suited for engineering design, which can be considered as a special class of inverse problems. Several examples from the research work of the author and his co-workers are presented and discussed to illustrate the main points raised in this paper.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.6
• /
• pp.2142-2153
• /
• 2014

Linearization of transducer characteristic plays a vital role in electronic instrumentation because all transducers have outputs nonlinearly related to the physical variables they sense. If the transducer output is nonlinear, it will produce a whole assortment of problems. Transducers rarely possess a perfectly linear transfer characteristic, but always have some degree of non-linearity over their range of operation. Attempts have been made by many researchers to increase the range of linearity of transducers. This paper presents a method to compensate nonlinearity of Linear Variable Displacement Transducer (LVDT) based on Extreme Learning Machine (ELM) method, Differential Evolution (DE) algorithm and Artificial Neural Network (ANN) trained by Genetic Algorithm (GA). Because of the mechanism structure, LVDT often exhibit inherent nonlinear input-output characteristics. The best approximation capability of optimized ANN technique is beneficial to this. The use of this proposed method is demonstrated through computer simulation with the experimental data of two different LVDTs. The results reveal that the proposed method compensated the presence of nonlinearity in the displacement transducer with very low training time, lowest Mean Square Error (MSE) value and better linearity. This research work involves less computational complexity and it behaves a good performance for nonlinearity compensation for LVDT and has good application prospect.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2004.11a
• /
• pp.244-249
• /
• 2004

There have been many attempts to predict the secondary structure content of a protein from its primary sequence, which serves as the first step in a series of bioinformatics processes to gain knowledge of the structure and function of a protein. Most of them assumed that prediction relying on the information of the amino acid composition of a protein can be successful. Several approaches expanded the amount of information by including the pair amino acid composition of two adjacent residues. Recent methods achieved a remarkable improvement in prediction accuracy by using this expanded composition information. The overall average errors of two successful methods were 6.1% and 3.4%. This work was motivated by the observation that evolutionarily related proteins share the similar structure. After manipulating the values of the frequency matrix obtained by running PSI-BLAST, inputs of an artificial neural network were constructed by taking the ratio of the amino acid composition of the evolutionarily related proteins with a query protein to the background probability. Although we did not utilize the expanded composition information of amino acid pairs, we obtained the comparable accuracy, with the overall average error being 3.6%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1951-1975
• /
• 2023

Recent advances in Cognitive Radio Networks (CRN) have elevated them to the status of a critical instrument for overcoming spectrum limits and achieving severe future wireless communication requirements. Collaborative spectrum sensing is presented for efficient channel selection because spectrum sensing is an essential part of CRNs. This study presents an innovative cooperative spectrum sensing (CSS) model that is built on the Firefly Algorithm (FA), as well as machine learning artificial neural networks (ANN). This system makes use of user grouping strategies to improve detection performance dramatically while lowering collaboration costs. Cooperative sensing wasn't used until after cognitive radio users had been correctly identified using energy data samples and an ANN model. Cooperative sensing strategies produce a user base that is either secure, requires less effort, or is faultless. The suggested method's purpose is to choose the best transmission channel. Clustering is utilized by the suggested ANN-FA model to reduce spectrum sensing inaccuracy. The transmission channel that has the highest weight is chosen by employing the method that has been provided for computing channel weight. The proposed ANN-FA model computes channel weight based on three sets of input parameters: PU utilization, CR count, and channel capacity. Using an improved evolutionary algorithm, the key principles of the ANN-FA scheme are optimized to boost the overall efficiency of the CRN channel selection technique. This study proposes the Artificial Neural Network with Firefly Algorithm (ANN-FA) for cognitive radio networks to overcome the obstacles. This proposed work focuses primarily on sensing the optimal secondary user channel and reducing the spectrum handoff delay in wireless networks. Several benchmark functions are utilized We analyze the efficacy of this innovative strategy by evaluating its performance. The performance of ANN-FA is 22.72 percent more robust and effective than that of the other metaheuristic algorithm, according to experimental findings. The proposed ANN-FA model is simulated using the NS2 simulator, The results are evaluated in terms of average interference ratio, spectrum opportunity utilization, three metrics are measured: packet delivery ratio (PDR), end-to-end delay, and end-to-average throughput for a variety of different CRs found in the network.