• The Transactions of the Korea Information Processing Society
• /
• v.5 no.11
• /
• pp.2897-2907
• /
• 1998

The production-rule generation from training examples is a hard problem that has large space and many local optimal solutions. Many learning methods are proposed for production-rule generation and genetic algorithms is an alternative learning method. However, traditional genetic algorithms has been known to have an obstacle in converging at the global solution area and show poor efficiency of production-rules generated. In this paper, we propose a production-rule generating method which uses genetic algorithm learning. By analyzing optimal sub-solutions captured by genetic algorithm learning, our method takes advantage of its schema structure and thus generates relatively small rule set.

• 대한공업교육학회지
• /
• v.42 no.1
• /
• pp.87-105
• /
• 2017

The purpose of this study was to verify that students' interest in team activities for problem solving, the effect of interest on it, and students' changes in perceptions and their behavioral characteristics in the process of problem solving. The study reviewed documents prepared by students, such as work sheets, descriptive questionnaires, works and their photos, student activity photos and observation journals of teachers. The results of this research are as below. First, a problem solving team activity for making a model car was considered an interesting assignment by more than 90% of male/female students. The fact that female students could be more focused on this assignment than male students was discovered. Interest in the assignment not only had an influence on the points from the start (the blueprint) to the end (model cars completed based upon the designs) of problem solving, but also provided the traction power behind the assignment. Second, the problem solving team activity allowed the students to change their existing recognition (thoughts) while positively taking a lead or indirectly utilizing various learning experiences (including experiences of failure). Third, $2^{nd}$ graders in middle school had a tendency to solve problems in dependently rather than to receive help from others when they encountered problematic situations.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.8
• /
• pp.11-18
• /
• 2009

This paper proposes the way of improving learning speed in Levenberg-Marquardt algorithm using the principal submatrix of Jacobian matrix. The Levenberg-Marquardt learning uses Jacobian matrix for Hessian matrix to get the second derivative of an error function. To make the Jacobian matrix an invertible matrix. the Levenberg-Marquardt learning must increase or decrease ${\mu}$ and recalculate the inverse matrix of the Jacobian matrix due to these changes of ${\mu}$. Therefore, to have the proper ${\mu}$, we create the principal submatrix of Jacobian matrix and set the ${\mu}$ as the eigenvalues sum of the principal submatrix. which can make learning speed improve without calculating an additional inverse matrix. We also showed that our method was able to improve learning speed in both a generalized XOR problem and a handwritten digit recognition problem.

• Journal of Korean Home Economics Education Association
• /
• v.22 no.1
• /
• pp.1-19
• /
• 2010

The purpose of this study was to develop practical problem focused teaching learning process plan for housing for the later life in order to apply it to the older stage of family planning section of Technology Home Economics in a highschool. Practical problem focused method was used for the teaching learning process plans of 3-session lessons according to the ADDIE model. The global practical problem was "What should I do to plan a safe and comfortable housing for the later life?" In the development stage, 53 teaching learning materials (44 students activity materials, 2 students' and 5 teacher' reading texts, and 2 moving pictures) were developed for 3-session lessons. The planes applied to the 5 classes, 150 students, in the freshmen of B highschool during April 20th-24th, 2009. The 5 point likert questionnaire were used to evaluate the 3-session lessons about 4 contents related aspects as well as the methods and effects of the lessons besides 2 open ended questions. The overall evaluation was very positive in all 6 aspects of the lessons. Some students wanted to learn more about universal design and aging related jobs. Those results showed that the practical problem focused teaching learning process plan for housing for the later life which this study developed would be appropriate to teach the older stage of family planning section related to housing and could be adjusted to the condition of each school and regions.

• Journal of KIISE:Software and Applications
• /
• v.29 no.12
• /
• pp.966-978
• /
• 2002

There are two important problems in improving text classification systems based on machine learning approach. The first one, called "selection problem", is how to select a minimum number of informative documents from a given document collection. The second one, called "composition problem", is how to reorganize selected training documents so that they can fit an adopted learning method. The former problem is addressed in "active learning" algorithms, and the latter is discussed in "boosting" algorithms. This paper proposes a new learning method, called AdaBUS, which proactively solves the above problems in the context of Naive Bayes classification systems. The proposed method constructs more accurate classification hypothesis by increasing the valiance in "weak" hypotheses that determine the final classification hypothesis. Consequently, the proposed algorithm yields perturbation effect makes the boosting algorithm work properly. Through the empirical experiment using the Routers-21578 document collection, we show that the AdaBUS algorithm more significantly improves the Naive Bayes-based classification system than other conventional learning methodson system than other conventional learning methods

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.20 no.2
• /
• pp.200-209
• /
• 2019

When forecasting future values, a model estimated after minimizing training errors can yield test errors higher than the training errors. This result is the over-fitting problem caused by an increase in model complexity when the model is focused only on a given dataset. Some regularization and resampling methods have been introduced to reduce test errors by alleviating this problem but have been designed for use with only a given dataset. In this paper, we propose a new optimization approach to reduce test errors by transforming a test error minimization problem into a training error minimization problem. To carry out this transformation, we needed additional data for the given dataset, termed pseudo data. To make proper use of pseudo data, we used three types of missing data imputation techniques. As an optimization tool, we chose the least squares method and combined it with an extra pseudo data instance. Furthermore, we present the numerical results supporting our proposed approach, which resulted in less test errors than the ordinary least squares method.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.3
• /
• pp.7-14
• /
• 2021

The emergence of the automated smart grid has become an essential device for responding to these problems and is bringing progress toward a smart grid-based society. Smart grid is a new paradigm that enables two-way communication between electricity suppliers and consumers. Smart grids have emerged due to engineers' initiatives to make the power grid more stable, reliable, efficient and safe. Smart grids create opportunities for electricity consumers to play a greater role in electricity use and motivate them to use electricity wisely and efficiently. Therefore, this study focuses on power demand management through machine learning. In relation to demand forecasting using machine learning, various machine learning models are currently introduced and applied, and a systematic approach is required. In particular, the GP learning model has advantages over other learning models in terms of general consumption prediction and data visualization, but is strongly influenced by data independence when it comes to prediction of smart meter data.

• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.1
• /
• pp.34-41
• /
• 2023

In recent years, as the range of human activities has increased, the introduction of alien species has become frequent. Among them, raccoons have been designated as harmful animals since 2020. Raccoons are similar in size and shape to raccoon dogs, so they generally need to be distinguished in capturing them. To solve this problem, we use VGG19, ResNet152V2, InceptionV3, InceptionResNet and NASNet, which are CNN deep learning models specialized for image classification. The parameters to be used for learning are pre-trained with a large amount of data, ImageNet. In order to classify the raccoon and raccoon dog datasets as outward features of animals, the image was converted to grayscale and brightness was normalized. Augmentation methods were applied using left and right inversion, rotation, scaling, and shift to create sufficient data for transfer learning. The FCL consists of 1 layer for the non-augmented dataset while 4 layers for the augmented dataset. Comparing the accuracy of various augmented datasets, the performance increased as more augmentation methods were applied.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.6
• /
• pp.183-193
• /
• 2021

In this paper, the author found and analyzed the problems caused by the change of traditional teaching methods to online in classrooms and laboratories. Looking at the analysis of major problems, first, there were various technical problems, including not all environments and facilities being connected to the Internet. Second, the effectiveness of virtual classes, which were suddenly switched online, could also be questioned. Finally, in the face of a new environment, the stress of teachers to adapt rapidly to the new teaching methodology was a problem. The author proposed digital transformation as a way to address these problems. The author analyzed educational changes, learning modalities and various technical tools, and various tasks to enable digital transformation. First, the author investigated, analyzed, and presented the factors necessary to efficiently operate the classroom environment that will change to online. Next, the author analyzed the factors and problems needed to make the students' classes reliable and efficient, and proposed solutions. Finally, the author pointed out the problem that during online lectures, the responsibility of learning is excessively transferred from teachers to students, and proposed a solution to this problem. Subsequently, the author proposed future studies.

• Journal of Elementary Mathematics Education in Korea
• /
• v.14 no.2
• /
• pp.263-285
• /
• 2010

It is necessary to accumulate the studies on the practical learning and teaching for the Mathematical talent education in elementary school. In this study, I set the 4th grade children mathematically gifted in elementary school to pose the various number calculating formulars, 4 4 4 4 = 0, 1, 2,$\cdots$10, by using to +, -, ${\times}$, $\div$, ( ). And I analysed their products. In 2007, I gave the same task to 5th graders and got a significant result. To expand the target of my study, I used the same investigating method for children of different graders. As a result, I conclude that math brains in 4th grade also can create various many number calculating formulas. I find that children pose to various many calaulating formulars becoming 0, 1, 8, 4 in order whereas they pose to a little calaulating formulars becoming 10, 6, 5, 9 orderly. Most errors are due to the order of calculation or confusion about parenthesis. This study contributes to test methods and text development for math brains in elementary school.