• Journal of KIISE:Software and Applications
• /
• v.32 no.8
• /
• pp.781-788
• /
• 2005

LDA (Linear Discriminant Analysis) is a data discrimination technique that seeks transformation to maximize the ratio of the between-class scatter and the within-class scatter While it has been successfully applied to several applications, it has two limitations, both concerning the underfitting problem. First, it fails to discriminate data with complex distributions since all data in each class are assumed to be distributed in the Gaussian manner; and second, it can lose class-wise information, since it produces only one transformation over the entire range of classes. We propose three extensions of LDA to overcome the above problems. The first extension overcomes the first problem by modeling the within-class scatter using a PCA mixture model that can represent more complex distribution. The second extension overcomes the second problem by taking different transformation for each class in order to provide class-wise features. The third extension combines these two modifications by representing each class in terms of the PCA mixture model and taking different transformation for each mixture component. It is shown that all our proposed extensions of LDA outperform LDA concerning classification errors for handwritten digit recognition and alphabet recognition.

• The Mathematical Education
• /
• v.51 no.3
• /
• pp.265-280
• /
• 2012

This study aims to provide implications from mathematics education perspective by designing a process of 'validity review on the problem solving process', and then, by analyzing the results. In the result of analysis on the features of children's thinking in accordance with 4 stages of problem solving, children's thinking was equally observed in every stage rather than intensively observed in one stage, and reflective thinking related to important elements from each stage of problem solving process was observed. In the result of analysis of changes in description for problem solving process, there was a difference in the aspects of changes by children's knowledge level in mathematics, however, the activity of validity review on problem solving process in overall induced positive changes in children's description, especially the changes in problem solving process of children. Through the result of this study, we could see that the validity review on problem solving process promotes children's reflective thinking and enables meta-cognition thus has a positive influence on children's description of problem solving process.

• The Journal of Information Systems
• /
• v.29 no.1
• /
• pp.75-91
• /
• 2020

Purpose Retrofit, defined to be addition of new technologies or features to the old system to increase efficiency or to abate GHG emissions, is considered as an important alternative for the old coal-fired power plant. The purpose of this study is to propose mathematical method to model multiple alternative retrofit in Generation Capacity Expansion Planning(GCEP) problem, and to get insight to the retrofit patterns from realistic case studies. Design/methodology/approach This study made a multi-alternative retrofit GECP model by adopting some new variables and equations to the existing GECP model. Added variables and equations are to ensure the retrofit feature that the life time of retrofitted plant is the remaining life time of the old power plant. We formulated such that multiple retrofit alternatives are simultaneously compared and the best retrofit alternative can be selected. And we found that old approach to model retrofit has a problem that old plant with long remaining life time is retrofitted earlier than the one with short remaining life time, fixed the problem by some constraints with some binary variables. Therefore, the proposed model is formulated into a mixed binary programming problem, and coded and run using the GAMS/cplex. Findings According to the empirical analysis result, we found that approach to model the multiple alternative retrofit proposed in this study is comparing simultaneously multiple retrofit alternatives and select the best retrofit satisfying the retrofit features related to the life time. And we found that retrofit order problem is cleared. In addition, the model is expected to be very useful in evaluating and developing the national policies concerning coal-fired power plant retrofit.

• Journal of Practical Engineering Education
• /
• v.14 no.1
• /
• pp.91-97
• /
• 2022

Recently, as per a representative education method to develop core capabilities (such as critical thinking, communication, collaboration, and creativity) problem based learning (PBL) has been widely adopted in universities. Two important features of PBL are 'collaboration between team members' and 'participation based self-directed learning'. These two features should be satisfied in online education, although it is difficult due to the limitation on space and time in the COVID-19 pandemic. This paper presents a new design of PBL class in Metaverse, based on improving the online PBL class operated in the previous semesters in the H university. In the proposed PBL class, students are able to display materials (e.g., image, pdf, video files) in 3D virtual space, that are related to problem solving. The 3D virtual space is called gallery in this paper. The concept of gallery allows for active participation of students. In addition, the gallery can be used as a tool for collaborative meeting or for final presentation. If possible, the new design of PBL class will be applied and its effectiveness will be analyzed.

• Management Science and Financial Engineering
• /
• v.13 no.2
• /
• pp.1-27
• /
• 2007

This paper presents a fuzzy-goal programming(FGP) approach for Bi-Level Linear Multiple Objective Decision Making(BLL-MODM) problem in a large hierarchical decision making and planning organization. The proposed approach combines the attractive features of both fuzzy set theory and goal programming(GP) for MODM problem. The GP problem has been developed by fixing the weights and aspiration levels for generating pareto-optimal(satisfactory) solution at each level for BLL-MODM problem. The higher level decision maker(HLDM) provides the preferred values of decision vector under his control and bounds of his objective function to direct the lower level decision maker(LLDM) to search for his solution in the right direction. Illustrative numerical example is provided to demonstrate the proposed approach.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1313-1318
• /
• 2017

Aggregating local features in a single vector is a fundamental problem in an image search. In this process, the image search process can be speeded up if binary features which are extracted almost two order of magnitude faster than gradient-based features are utilized. However, in order to utilize the binary features in an image search, it is necessary to study the techniques for clustering binary features to generate binary visual words. This investigation is necessary because traditional clustering techniques for gradient-based features are not compatible with binary features. To this end, this paper studies the techniques for clustering binary features for the purpose of generating binary visual words. Through experiments, we analyze the trade-off between the accuracy and computational efficiency of an image search using binary features, and we then compare the proposed techniques. This research is expected to be applied to mobile applications, real-time applications, and web scale applications that require a fast image search.

• Journal of KIISE:Software and Applications
• /
• v.33 no.8
• /
• pp.718-729
• /
• 2006

There has been increasing interest in automatic term recognition (ATR), which recognizes technical terms for given domain specific texts. ATR is composed of 'term extraction', which extracts candidates of technical terms and 'term selection' which decides whether terms in a term list derived from 'term extraction' are technical terms or not. 'term selection' is a process to rank a term list depending on features of technical term and to find the boundary between technical term and general term. The previous works just use statistical features of terms for 'term selection'. However, there are limitations on effectively selecting technical terms among a term list using the statistical feature. The objective of this paper is to find effective features for 'term selection' by considering various aspects of technical terms. In order to solve the ranking problem, we derive various features of technical terms and combine the features using machine-learning algorithms. For solving the boundary finding problem, we define it as a binary classification problem which classifies a term in a term list into technical term and general term. Experiments show that our method records 78-86% precision and 87%-90% recall in boundary finding, and 89%-92% 11-point precision in ranking. Moreover, our method shows higher performance than the previous work's about 26% in maximum.

• Journal of Information Management
• /
• v.41 no.2
• /
• pp.31-46
• /
• 2010

Now named-entity recognition(NER) as a part of information extraction has been used in the fields of information retrieval as well as question-answering systems. Unlike words, named-entities(NEs) are generated and changed steadily in documents on the Web, newspapers, and so on. The NE generation causes an unknown word problem and makes many application systems with NER difficult. In order to alleviate this problem, this paper proposes a new feature generation method for machine learning-based NER. In general features in machine learning-based NER are related with words, but entities in named-entity dictionaries are related to phrases. So the entities are not able to be directly used as features of the NER systems. This paper proposes an encoding scheme as a feature generation method which converts phrase entities into features of word units. Futhermore, due to this scheme, entities with semantic information in WordNet can be converted into features of the NER systems. Through our experiments we have shown that the performance is increased by about 6% of F1 score and the errors is reduced by about 38%.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.8
• /
• pp.81-87
• /
• 1999

In general, feature detection and tracking algorithms is classified by EBGM using Garbor-jet, NNC-R and STK algorithm using pixel eigenvalue. In those algorithms, EBGM and NCC-R detect features with feature model, but STK algorithm has a characteristics of an automatic feature selection. In this paper, to solve the initial problem of NR tracking in STK algorithm, we detected features using STK algorithm in modelled feature region and tracked features with NR method. In tracking, to improve the tracking accuracy for features by NR method, we proposed BMA-NR method. We evaluated that BMA-NR method was superior to NBMA-NR in that feature tracking accuracy, since BMA-NR method was able to solve the local minimum problem due to search window size of NR.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.3
• /
• pp.144-154
• /
• 2003

In this paper we present Logical Evolution method which is a new teaming algorithm for the concepts expressed as binary logic function. We try to solve some problems of Inductive Learning algorithms through Logical Evolution. First, to be less affected from limited prior knowledge, it generates features using the gained informations during learning process and learns the concepts with these features. Second, the teaming is done using not the whole example set but the individual example, so even if new problem or new input-output variables are given, it can use the previously generated features. In some cases these old features can make the teaming process more efficient. Logical Evolution method consists of 5 operations which are selected and performed by the logical evaluation procedure for feature generation and learning process. To evaluate the performance of the present algorithm, we make experiments on MONK data set and a newly defined problem.