• Journal of the Korean Data and Information Science Society
• /
• v.22 no.1
• /
• pp.99-105
• /
• 2011

We propose a semi-supervised learning algorithm based on a form of regularization that incorporates similarity and dissimilarity penalty terms. Our approach uses a graph-based encoding of similarity and dissimilarity. We also present a model-selection method which employs cross-validation techniques to choose hyperparameters which affect the performance of the proposed method. Simulations using two types of dat sets demonstrate that the proposed method is promising.

• Journal of The Korean Association For Science Education
• /
• v.28 no.8
• /
• pp.796-813
• /
• 2008

A good understanding of how gifted science students understand physics is important to developing and delivering effective curriculum for gifted science students. This dissertation reports on a systematic investigation of gifted science students' reasoning model in learning physics. An analysis of videotaped class work, written work and interviews indicate that I will discuss the framework to characterize student reasoning. There are three main groups of students. The first group of gifted science students holds several different understandings of a single concept and apply them inconsistently to the tasks related to that concept. Most of these students hold the Aristotelian Model about Newton's second law. In this case, I define this reasoning model as the manifold model. The second group of gifted science students hold a unitary understanding of a single concept and apply it consistently to several tasks. Most of these students hold a Newtonian Model about Newton's second law. In this case, I define this reasoning model as the coherence model. Finally, some gifted science students have a manifold model with several different perceptions of a single concept and apply them inconsistently to tasks related to the concept. Most of these students hold the Aristotelian Model about Newton's second law. In this case, I define this reasoning model as the coherence model.

• Journal of Biomedical Engineering Research
• /
• v.45 no.4
• /
• pp.179-186
• /
• 2024

This paper investigates the impact of Riemannian Procrustes Analysis (RPA) on enhancing the classification performance of SPD-Net when applied to EEG signals across different sessions and subjects. EEG signals, known for their inherent individual variability, are initially transformed into Symmetric Positive Definite (SPD) matrices, which are naturally represented on a Riemannian manifold. To mitigate the variability between sessions and subjects, we employ RPA, a method that geometrically aligns the statistical distributions of these matrices on the manifold. This alignment is designed to reduce individual differences and improve the accuracy of EEG signal classification. SPD-Net, a deep learning architecture that maintains the Riemannian structure of the data, is then used for classification. We compare its performance with the Minimum Distance to Mean (MDM) classifier, a conventional method rooted in Riemannian geometry. The experimental results demonstrate that incorporating RPA as a preprocessing step enhances the classification accuracy of SPD-Net, validating that the alignment of statistical distributions on the Riemannian manifold is an effective strategy for improving EEG-based BCI systems. These findings suggest that RPA can play a role in addressing individual variability, thereby increasing the robustness and generalization capability of EEG signal classification in practical BCI applications.

• Journal of Korea Multimedia Society
• /
• v.17 no.1
• /
• pp.34-42
• /
• 2014

In this paper an embedding algorithm based on commute time is implemented by organizing patches according to the graph-based metric, and its performance is analyzed by comparing with the results of principal component analysis embedding. It is usual that the dimensionality reduction be done within some acceptable approximation error. However this paper shows the proposed manifold embedding method generates the intrinsic geometry corresponding to the signal despite severe approximation error, so that it can be applied to the areas such as pattern classification or machine learning.

• Journal of the Korean Institute of Educational Facilities
• /
• v.12 no.5
• /
• pp.13-24
• /
• 2005

This study is the second that aims at offering the basic information on the appropriate spatial organization of the special classes by looking at the relationship between a group of learning activities and a group of playing activities in Elementary and Middle Schools The learning space unit of the special classes should be more flexible for the various learning activities and be prepared in order to correspond to the needs of a territory for different learning appeared according to the degree of handicap, learning ability and the contents of learning. This study dealt with the learning space unit to tackle the problems of special classes. In fact, it is unwise to offer so many different kinds of learning spaces in every school. Due to the manifold and multiple characteristics of handicap, the problem of special classes should be approached by the overall educational system of special educational facilities rather than by a special classes space alone. In this respect, it can be said that this problem should be tackled by reorganization of the special classes in the community through specialization and network system of special class facilities in order to make more effective educational environment.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.103.1-103
• /
• 2002

In variable structure control algorithms, The control law used to realized the desired sliding mode dynamics is discontinuous on the switching manifold. However, due to imperfections in switching, such as time delays, the system trajectory chatters instead of sliding along the switching manifold. This chattering is undesirable because it may excite unmodeled high frequency dynamics in the physical system. In this paper, to overcome this drawback a self-organizing fuzzy sliding mode control algorithm using gradient descent method is proposed. The proposed method has the characteristics which are viewed in conventional VSC, e.g. insensitivity to a class of disturbance, parameter variations and uncertainties ill the sliding mode. To demonstrate its performance, the proposed control algorithm is applied to an inverted pendulum system. The results show that both alleviation of chattering and performance are achieved.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1721-1737
• /
• 2020

Human apparent age estimation has become a popular research topic and attracted great attention in recent years due to its wide applications, such as personal security and law enforcement. To achieve the goal of age estimation, a large number of methods have been pro-posed, where the models derived through the cumulative attribute coding achieve promised performance by preserving the neighbor-similarity of ages. However, these methods afore-mentioned ignore the geometric structure of extracted facial features. Indeed, the geometric structure of data greatly affects the accuracy of prediction. To this end, we propose an age estimation algorithm through joint feature selection and manifold learning paradigms, so-called Feature-selected and Geometry-preserved Least Square Regression (FGLSR). Based on this, our proposed method, compared with the others, not only preserves the geometry structures within facial representations, but also selects the discriminative features. Moreover, a deep learning extension based FGLSR is proposed later, namely Feature selected and Geometry preserved Neural Network (FGNN). Finally, related experiments are conducted on Morph2 and FG-Net datasets for FGLSR and on Morph2 datasets for FGNN. Experimental results testify our method achieve the best performances.

• Journal of the Korean Institute of Educational Facilities
• /
• v.12 no.4
• /
• pp.17-29
• /
• 2005

The purpose of this study is to establish fundamental standards of architectural planning concerning special class facilities in order to offer the basic information on the appropriate spatial organization of the special classroom by looking at the relationship between learning activities and living activities and the existing spatial organization. At present, there are no proper architectural standards which correspond to special class children's handicap and it's various characteristics. The special classes are just using ordinary classrooms without a considerations of the children with manifold handicap. In this sense, this study deals with appropriate special class facilities corresponding to the various characteristics of children's handicap, the contacting activities of special children with ordinary children and finally proper environment for the mainstreaming education which special education pursues.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10a
• /
• pp.211-212
• /
• 2007

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.300-303
• /
• 2003

An Algorithm based on direct implementation of variable structure systems theory approach is proposed for on-line training of multilayer perceptrons. Network structures which have multiple inputs, single output and one hidden layer are considered and the weights are assumed to have capabilities for continuous time adaptation. The zero level set of the network learning error is regarded as a sliding surface in the learning parameters space. A sliding mode trajectory can be brought on and reached in finite time on such a sliding manifold. Results from simulated on-line identification task for a two-link planar manipulator dynamics are also presented.