• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.10a
• /
• pp.433-436
• /
• 2004

In this paper, we introduce a new Fuzzy Polynomial Neural Networks (FPNNS)-like structure whose neuron is based on the Fuzzy Set-based Fuzzy Inference System (FS-FIS) and is different from that of FPNNS based on the Fuzzy relation-based Fuzzy Inference System (FR-FIS) and discuss the ability of the new FPNNS-like structure named Fuzzy Set-based Polynomial Neural Networks (FSPNN). The premise parts of their fuzzy rules are not identical, while the consequent parts of the both Networks (such as FPNN and FSPNN) are identical. This difference results from the angle of a viewpoint of partition of input space of system. In other word, from a point of view of FS-FIS, the input variables are mutually independent under input space of system, while from a viewpoint of FR-FIS they are related each other. The proposed design procedure for networks architecture involves the selection of appropriate nodes with specific local characteristics such as the number of input variables, the order of the polynomial that is constant, linear, quadratic, or modified quadratic functions being viewed as the consequent part of fuzzy rules, and a collection of the specific subset of input variables. On the parameter optimization phase, we adopt Information Granulation (IC) based on HCM clustering algorithm and a standard least square method-based learning. Through the consecutive process of such structural and parametric optimization, an optimized and flexible fuzzy neural network is generated in a dynamic fashion. To evaluate the performance of the genetically optimized FSPNN (gFSPNN), the model is experimented with using the time series dataset of gas furnace process.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.8
• /
• pp.232-237
• /
• 2013

A newly developed fall detection algorithm using the HMM (Hidden Markov Model) extracted from the video is introduced. To distinguish between the fall from personal difference fall pattern or the normal activities of daily living (ADL), HMM machine learning algorithm is used. For getting fall feature vector of video, the motion vector from the optical flow is applied to the PCA (Principal Component Analysis). The combination of the angle, ratio of long-short axis, velocity from results of PCA make the new fall feature parameters. These parameters were applied to the HMM and the results were compared and analyzed. Among the newly proposed various kinds of fall parameters, the angle of movement showed the best results. The results show that this parameter can distinguish various types of fall from ADLs with 91.5% sensitivity and 88.01% specificity.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.9
• /
• pp.719-726
• /
• 2020

Education policy is a dynamic process featuring social development trends. The world countries have focused their education program on empowering the learners for future life and work. This paper aims to assess the higher education curriculum based on a survey of 280 students, employers, alumni, and lecturers in both social sciences and natural sciences in Ho Chi Minh City, Vietnam. The fuzzy decision-making method, namely the Fuzzy Extent Analysis Method (F-EAM), was applied to measure the relative weight of each parameter. Seven factors under the curriculum development have been put in the ranking. Input with emphasis on foreign language was the highest priority in curriculum development, given the expected demand of the labor market. Objective and learning outcome and teaching activities ranked second and third, respectively. The traditional triangle of teaching content, methodology, and evaluation and assessment are still proven their roles, but certain modifications have been defined in the advanced curriculum. Teaching facilities had the least weight among the seven dimensions of curriculum development. The findings are helpful for education managers to efficiently allocate scarce resources to reform the curriculum to bridge the undergraduate quality gap between labor supply and demand, meeting the dynamic trends of social development.

• The Journal of the Acoustical Society of Korea
• /
• v.11 no.3
• /
• pp.61-66
• /
• 1992

There have been many researches that uses neural network models for automatic speech recognition, but the main trend was finding the neural network models and learning rules appropriate to automatic speech recognition. However, the choice of the input speech parameter for the neural network as well as neural network model itself is a very important factor for the improvement of performance of the automatic speech recognition system using neural network. In this paper we select 6 speech parameters from surveys of the speech recognition papers which uses neural networks, and analyze the performance for the same data and the same neural network model. We use 8 sets of 9 Korean plosives and 18 sets of 8 Korean vowels. We use recurrent neural network and compare the performance of the 6 speech parameters while the number of nodes is constant. The delta cepstrum of linear predictive coefficients showed best result and the recognition rates are 95.1% for the vowels and 100.0% for plosives.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.8
• /
• pp.2188-2195
• /
• 1999

This paper proposes a new call admission control scheme utilizing an inverse fuzzy vector quantizer(IFVQ) and neuralnet, which combines benefits of IFVQ and flexibilities of FCM(Fuzzy-C-Means) arithmetics, to decide whether a requested call not to be trained in learning phase to be connected or not. The system generates the estimated traffic pattern for the cell stream of a new call, using feasible/infeasible patterns in codebook, fuzzy membership values that represent the degree to which each pattern of codebook matches input pattern, and FCM arithmetics. The input to the NN is the vector consisted of traffic parameters which are the means and variances of the number of cells arriving in decision as to whether to accept or reject a new call depends on whether the NN is used for decision threshold(+0.5). This method is a new technique for call admission control using the membership values as traffic parameter which declared to CAC at the call set up stage, and this is valid for a very general traffic model in which the calls of a stream can belong to an unlimited number of traffic classes. Through the simulations, it is founded the performance of the suggested method outperforms compared to the conventional NN method.

• Journal of Power Electronics
• /
• v.13 no.4
• /
• pp.516-527
• /
• 2013

This paper presents a new overall system for state-of-available-power (SoAP) prediction for a lithium-ion battery pack. The essential part of this method is based on an adaptive network architecture which utilizes both fuzzy model (FIS) and artificial neural network (ANN) into the framework of adaptive neuro-fuzzy inference system (ANFIS). While battery aging proceeds, the system is capable of delivering accurate power prediction not only for room temperature, but also at lower temperatures at which power prediction is most challenging. Due to design property of ANN, the network parameters are adapted on-line to the current battery states (state-of-charge (SoC), state-of-health (SoH), temperature). SoC is required as an input parameter to SoAP module and high accuracy is crucial for a reliable on-line adaptation. Therefore, a reasonable way to determine the battery state variables is proposed applying a combination of several partly different algorithms. Among other SoC boundary estimation methods, robust extended Kalman filter (REKF) for recalibration of amp hour counters was implemented. ANFIS then achieves the SoAP estimation by means of time forward voltage prognosis (TFVP) before a power pulse occurs. The trade-off between computational cost of batch-learning and accuracy during on-line adaptation was optimized resulting in a real-time system with TFVP absolute error less than 1%. The verification was performed on a software-in-the-loop test bench setup using a 53 Ah lithium-ion cell.

• Transactions of Materials Processing
• /
• v.29 no.5
• /
• pp.241-250
• /
• 2020

In this study, the methodology of the springback prediction of automotive parts applied 3rd generation AHSS was investigated using the response surface model analysis based on a regression model, and the meta model analysis based on a Kriging model. To design the learning data set for constructing the springback prediction models, and the experimental design was conducted at three levels for each processing variable using the definitive screening designs method. The hat-shaped member, which is the basic shape of the member parts, was selected and the springback values were measured for each processing type and processing variable using the finite element analysis. When the nonlinearity of the variables is small during the hat-shaped member forming, the response surface model and the meta model can provide the same processing parameter. However, the accuracy of the springback prediction of the meta model is better than the response surface model. Even in the case of the simple shape parts forming, the springback prediction accuracy of the meta model is better than that of the response surface model, when more variables are considered and the nonlinearity effect of the variables is large. The efficient global optimization algorithm-based Kriging is appropriate in resolving the high computational complexity optimization problems such as developing automotive parts.

• Phonetics and Speech Sciences
• /
• v.12 no.3
• /
• pp.55-63
• /
• 2020

Voice conversion can be applied to various voice processing applications. It can also play an important role in data augmentation for speech recognition. The conventional method uses the architecture of voice conversion with speech synthesis, with Mel filter bank as the main parameter. Mel filter bank is well-suited for quick computation of neural networks but cannot be converted into a high-quality waveform without the aid of a vocoder. Further, it is not effective in terms of obtaining data for speech recognition. In this paper, we focus on performing voice-to-voice conversion using only the raw spectrum. We propose a deep learning model based on the transformer network, which quickly learns the voice conversion properties using an attention mechanism between source and target spectral components. The experiments were performed on TIDIGITS data, a series of numbers spoken by an English speaker. The conversion voices were evaluated for naturalness and similarity using mean opinion score (MOS) obtained from 30 participants. Our final results yielded 3.52±0.22 for naturalness and 3.89±0.19 for similarity.

• Advances in environmental research
• /
• v.5 no.3
• /
• pp.153-167
• /
• 2016

We applied multilayer perceptron (MLP) and radial basis function (RBF) neural network in upstream and downstream water quality stations of the Karaj Reservoir in Iran. For both neural networks, inputs were pH, turbidity, temperature, chlorophyll-a, biochemical oxygen demand (BOD) and nitrate, and the output was dissolved oxygen (DO). We used an MLP neural network with two hidden layers, for upstream station 15 and 33 neurons in the first and second layers respectively, and for the downstream station, 16 and 21 neurons in the first and second hidden layer were used which had minimum amount of errors. For learning process 6-fold cross validation were applied to avoid over fitting. The best results acquired from RBF model, in which the mean bias error (MBE) and root mean squared error (RMSE) were 0.063 and 0.10 for the upstream station. The MBE and RSME were 0.0126 and 0.099 for the downstream station. The coefficient of determination ($R^2$) between the observed data and the predicted data for upstream and downstream stations in the MLP was 0.801 and 0.904, respectively, and in the RBF network were 0.962 and 0.97, respectively. The MLP neural network had acceptable results; however, the results of RBF network were more accurate. A sensitivity analysis for the MLP neural network indicated that temperature was the first parameter, pH the second and nitrate was the last factor affecting the prediction of DO concentrations. The results proved the workability and accuracy of the RBF model in the prediction of the DO.

• Journal of Radiation Protection and Research
• /
• v.42 no.4
• /
• pp.212-221
• /
• 2017

Background: This study aims to calculate detector positions as a design of a radioactive source localizing radiation portal monitor (RPM) system using an improved genetic algorithm. Materials and Methods: To calculate of detector positions for a source localizing RPM system optimization problem is defined. To solve the problem, a modified iterative genetic algorithm (MIGA) is developed. In general, a genetic algorithm (GA) finds a globally optimal solution with a high probability, but it is not perfect at all times. To increase the probability to find globally optimal solution rather, a MIGA is designed by supplementing the iteration, competition, and verification with GA. For an optimization problem that is defined to find detector positions that maximizes differences of detector signals, a localization method is derived by modifying the inverse radiation transport model, and realistic parameter information is suggested. Results and Discussion: To compare the MIGA and GA, both algorithms are implemented in a MATLAB environment. The performance of the GA and MIGA and that of the procedures supplemented in the MIGA are analyzed by computer simulations. The results show that the iteration, competition, and verification procedures help to search for globally optimal solutions. Further, the MIGA is more robust against falling into local minima and finds a more reliably optimal result than the GA. Conclusion: The positions of the detectors on an RPM for radioactive source localization are optimized using the MIGA. To increase the contrast of the measurements from each detector, a relationship between the source and the detectors is derived by modifying the inverse transport model. Realistic parameters are utilized for accurate simulations. Furthermore, the MIGA is developed to achieve a reliable solution. By utilizing results of this study, an RPM for radioactive source localization has been designed and will be fabricated soon.