• 한국IT서비스학회지
• /
• 제9권1호
• /
• pp.113-125
• /
• 2010

Recently, a smart space with Ubiquitous Environment is expanding rapidly due to the development of Ubiquitous Sensor Network. Therefore, more appropriate and intelligent services of the context awareness system is being required. The previous context awareness system can provide a service to the user through the inference only on the current situation. But, it does not handle certain situation properly when the system provides abnormal result. Also it does not have any proper method of generating reliable semantic data from sensed raw data. In this paper, we are trying to solve the problems as the following approaches. First, the system recognizes abnormal result and corrects it by learning feedback from the user. Second, we suggest a method of converting sensed data into more reliable semantic data. Third, we build the system based on an Ontological context model that is capable of interoperability and reusability. Therefore, the context awareness system of our study can enhance the previous system that can generate more reliable context data, can provide more effective inference method, and can provide more intelligent system structure.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.366-366
• /
• 2000

In this paper, an adaptive neuro-fuzzy filter using the conditional fuzzy c-means(CFCM) methods is proposed. Usualy, the number of fuzzy rules exponentially increases by applying the grid partitioning of the input space, in conventional adaptive neuro-fuzzy inference system(ANFIS) approaches. In order to solve this problem, CFCM method is adopted to render the clusters which represent the given input and output data. Parameter identification is performed by hybrid learning using back-propagation algorithm and total least square(TLS) method. Finally, we applied the proposed method to the nonlinear channel equalization problem and obtained a better performance than previous works.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1994년도 추계학술대회 논문집
• /
• pp.414-419
• /
• 1994

In recent years, there has been an increasing intest in control of active automotive suspension systems with a goal of improving the ride comfort and safety. Many approaches for these purposes have used linearized models of the suspension's dynamics, allowing the use of linear control theory. However, the linearized model does not well descriibe the actual system behavior which is inherently nonlinear. The object of this study is to develop a neuro controlled active suspension for the ride quality improvement. After obtaining active control law using optimal control theory, we use the artificial neural network to train the neuro controller to learn the relation of road input and control force. Form the numerical results, we found that back propagation learning does show good pattern matching and vertical acceleration of the driver's seat and sprung mass.

• 한국환경교육학회지:환경교육
• /
• 제22권3호
• /
• pp.112-124
• /
• 2009

Visual images are important elements of the communication of environment in secondary school environment textbooks, which dominate classroom approaches to teaching and learning. This study was to analysed the perspectives of pedagogical and social-pedagogic of visual images in 'global warming' unit of six different Korean environment textbooks. Analysis makes use of the dimensions pedagogical functions, content specialization, social-pedagogic relationships, and abstraction of the visual codes. The results showed that most of the visual images in school environment textbooks tend to establish social-pedagogic relationships with students. The results of our study have implications for textbook authors and secondary environment teachers. We suggest that future studies may focus on students' and teacher's interpretation of visual images new secondary environment textbooks.

• ETRI Journal
• /
• 제39권6호
• /
• pp.832-840
• /
• 2017

Recently, deep recurrent neural networks have achieved great success in various machine learning tasks, and have also been applied for sound event detection. The detection of temporally overlapping sound events in realistic environments is much more challenging than in monophonic detection problems. In this paper, we present an approach to improve the accuracy of polyphonic sound event detection in multichannel audio based on gated recurrent neural networks in combination with auditory spectral features. In the proposed method, human hearing perception-based spatial and spectral-domain noise-reduced harmonic features are extracted from multichannel audio and used as high-resolution spectral inputs to train gated recurrent neural networks. This provides a fast and stable convergence rate compared to long short-term memory recurrent neural networks. Our evaluation reveals that the proposed method outperforms the conventional approaches.

• 한국지능시스템학회논문지
• /
• 제12권3호
• /
• pp.261-268
• /
• 2002

This paper represents that auto tunings of membership functions and weights in the fuzzy neural networks are effectively performed by immune algorithm. A number of hybrid methods in fuzzy-neural networks are considered in the context of tuning of learning method, a general view is provided that they are the special cases of either the membership functions or the gain modification in the neural networks by genetic algorithms. On the other hand, since the immune network system possesses a self organizing and distributed memory, it is thus adaptive to its external environment and allows a PDP (parallel distributed processing) network to complete patterns against the environmental situation. Also, it can provide optimal solution. Simulation results reveal that immune algorithms are effective approaches to search for optimal or near optimal fuzzy rules and weights.

• 전기학회논문지
• /
• 제67권6호
• /
• pp.767-772
• /
• 2018

Recently, some meta-heuristic algorithms, such as GA(Genetic Algorithm) and GP(Genetic Programming), have been used to optimize CNN(Convolutional Neural Network). The CNN, which is one of the deep learning models, has seen much success in a variety of computer vision tasks. However, designing CNN architectures still requires expert knowledge and a lot of trial and error. In this paper, the recent attempts to automatically construct CNN architectures are investigated and analyzed. First, two GA based methods are summarized. One is the optimization of CNN structures with the number and size of filters, connection between consecutive layers, and activation functions of each layer. The other is an new encoding method to represent complex convolutional layers in a fixed-length binary string, Second, CGP(Cartesian Genetic Programming) based method is surveyed for CNN structure optimization with highly functional modules, such as convolutional blocks and tensor concatenation, as the node functions in CGP. The comparison for three approaches is analysed and the outlook for the potential next steps is suggested.

• 한국수학교육학회지시리즈D:수학교육연구
• /
• 제14권4호
• /
• pp.347-360
• /
• 2010

Motivating students to study mathematics is a concern for many mathematics educators. In this paper, we present the outcome of a one semester experiment in which group-work and academic advising were used as teaching strategies in order to improve the motivational level of our students in learning mathematics. Although the students' performance did not show any statistically significant difference between the experimental and control groups, qualitative and other quantitative data collected indicate that the participants in the experiment, especially weak students, have in one way or the other benefited from the teaching approaches. Details of the experiment, the findings and their educational implications are presented.

• 전기학회논문지
• /
• 제67권12호
• /
• pp.1665-1670
• /
• 2018

A convolutional neural network (CNN), which is one of the deep learning models, has been very successful in a variety of computer vision tasks. Filters of a CNN are automatically generated, however, they can be further optimized since there exist the possibility of existing redundant and less important features. Therefore, the aim of this paper is a filter reduction to accelerate and compress CNN models. Evolutionary algorithms is adopted to remove the unnecessary filters in order to minimize the parameters of CNN networks while maintaining a good performance of classification. We demonstrate the proposed filter reduction methods performing experiments on CIFAR10 data based on the classification performance. The comparison for three approaches is analysed and the outlook for the potential next steps is suggested.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2013년도 춘계학술발표대회
• /
• pp.204-207
• /
• 2013

Music genre classification is an essential component for music information retrieval system. There are two important components to be considered for better genre classification, which are audio feature extraction and classifier. This paper incorporates two different kinds of features for genre classification, timbral texture and rhythmic content features. Timbral texture contains several spectral and Mel-frequency Cepstral Coefficient (MFCC) features. Before choosing a timbral feature we explore which feature contributes less significant role on genre discrimination. This facilitates the reduction of feature dimension. For the timbral features up to the 4-th order central moments and the covariance components of mutual features are considered to improve the overall classification result. For the rhythmic content the features extracted from beat histogram are selected. In the paper Extreme Learning Machine (ELM) with bagging is used as classifier for classifying the genres. Based on the proposed feature sets and classifier, experiment is performed with well-known datasets: GTZAN databases with ten different music genres, respectively. The proposed method acquires the better classification accuracy than the existing approaches.