• Journal of Information Technology Services
• /
• v.18 no.2
• /
• pp.143-159
• /
• 2019

Machine learning (ML) is a method of fitting given data to a mathematical model to derive insights or to predict. In the age of big data, where the amount of available data increases exponentially due to the development of information technology and smart devices, ML shows high prediction performance due to pattern detection without bias. The feature engineering that generates the features that can explain the problem to be solved in the ML process has a great influence on the performance and its importance is continuously emphasized. Despite this importance, however, it is still considered a difficult task as it requires a thorough understanding of the domain characteristics as well as an understanding of source data and the iterative procedure. Therefore, we propose methods to apply deep learning for solving the complexity and difficulty of feature extraction and improving the performance of ML model. Unlike other techniques, the most common reason for the superior performance of deep learning techniques in complex unstructured data processing is that it is possible to extract features from the source data itself. In order to apply these advantages to the business problems, we propose deep learning based methods that can automatically extract features from transaction data or directly predict and classify target variables. In particular, we applied techniques that show high performance in existing text processing based on the structural similarity between transaction data and text data. And we also verified the suitability of each method according to the characteristics of transaction data. Through our study, it is possible not only to search for the possibility of automated feature extraction but also to obtain a benchmark model that shows a certain level of performance before performing the feature extraction task by a human. In addition, it is expected that it will be able to provide guidelines for choosing a suitable deep learning model based on the business problem and the data characteristics.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.45 no.1
• /
• pp.28-45
• /
• 2021

The importance of smart clothing as a product is increasingly emphasized as further growth in the potential of the smart market is expected. There is a high understanding and sympathy for the potential of smart clothing in the mass consumer market; therefore, commercialization is not actively carried out. This study enhances the understanding of the development direction of products with a focus on technical benefits, in order for smart clothing to gain access to customers as wearable devices. This study identifies major technologies used in smart clothing through an analysis of the patent technology status of smart clothing in Korea. Smart clothing is divided into three types: passive smart, active smart and advanced smart clothing based on a reaction mechanism and functional scope. We present the smart clothing and discuss the product features for three types. According to research, smart clothing products were equipped with passive, active, and advanced smart systems as well as provided new services by converging big data and AI technologies, rather than only using technologies such as sensors, controls, and actuators. Future directions for new smart clothing product development is also discussed in the conclusion.

• The Mathematical Education
• /
• v.62 no.3
• /
• pp.435-455
• /
• 2023

This study primarily focused on the development of an Explainable Artificial Intelligence (XAI) model to discern and analyze papers with significant impact in the field of mathematics education. To achieve this, meta-information from 29 domestic and international mathematics education journals was utilized to construct a comprehensive academic research network in mathematics education. This academic network was built by integrating five sub-networks: 'paper and its citation network', 'paper and author network', 'paper and journal network', 'co-authorship network', and 'author and affiliation network'. The Random Forest machine learning model was employed to evaluate the impact of individual papers within the mathematics education research network. The SHAP, an XAI model, was used to analyze the reasons behind the AI's assessment of impactful papers. Key features identified for determining impactful papers in the field of mathematics education through the XAI included 'paper network PageRank', 'changes in citations per paper', 'total citations', 'changes in the author's h-index', and 'citations per paper of the journal'. It became evident that papers, authors, and journals play significant roles when evaluating individual papers. When analyzing and comparing domestic and international mathematics education research, variations in these discernment patterns were observed. Notably, the significance of 'co-authorship network PageRank' was emphasized in domestic mathematics education research. The XAI model proposed in this study serves as a tool for determining the impact of papers using AI, providing researchers with strategic direction when writing papers. For instance, expanding the paper network, presenting at academic conferences, and activating the author network through co-authorship were identified as major elements enhancing the impact of a paper. Based on these findings, researchers can have a clear understanding of how their work is perceived and evaluated in academia and identify the key factors influencing these evaluations. This study offers a novel approach to evaluating the impact of mathematics education papers using an explainable AI model, traditionally a process that consumed significant time and resources. This approach not only presents a new paradigm that can be applied to evaluations in various academic fields beyond mathematics education but also is expected to substantially enhance the efficiency and effectiveness of research activities.

• Wind and Structures
• /
• v.9 no.6
• /
• pp.441-455
• /
• 2006

A more applicable optimization model for extracting flutter derivatives of bridge decks is presented, which is suitable for time-varying weights for fitting errors and different lengths of vertical bending and torsional free vibration data. A stochastic search technique for searching the optimal solution of optimization problem is developed, which is more convenient in understanding and programming than the alternate iteration technique, and testified to be a valid and efficient method using two numerical examples. On the basis of the section model test of Sutong Bridge deck, the flutter derivatives are extracted by the stochastic search technique, and compared with the identification results using the modified least-square method. The Empirical Mode Decomposition method is employed to eliminate noise, trends and zero excursion of the collected free vibration data of vertical bending and torsional motion, by which the identification precision of flutter derivatives is improved.

• Asian-Australasian Journal of Animal Sciences
• /
• v.22 no.11
• /
• pp.1477-1486
• /
• 2009

Female germ cell meiotic divisions are typically asymmetric, giving rise to two daughter cells with different sizes. Spindle movements including spindle migration from the oocyte center to the cortex and spindle rotation from parallel to perpendicular (typically in the mouse) at the cortex are crucial for these asymmetric divisions and therefore are crucial for gamete production. Different regulatory mechanisms for spindle movements have been determined in different species and a wide variety of different molecular components and processes that are involved in spindle movements have also been identified in different species. Here, we review the current state of knowledge as well as our understanding of mechanisms for spindle movements in different systems with focus on three main aspects: microtubules (MT), microfilaments (MF) and molecules associated with cytoskeletal organization as well as molecules that are not directly related to the cytoskeleton. How they might interact or function independently during female meiotic divisions in different species is discussed in detail.

• Food Science and Biotechnology
• /
• v.15 no.6
• /
• pp.889-895
• /
• 2006

Cytotoxic effects of extracts from Tremella fuciformis strain FB001 were evaluated on the DLD-1 human colon adenocarcinoma cell line and the content of polyphenolic compounds in the extracts were analyzed. Hexane, chloroform, and ethyl acetate subfractions (experimental setting I) exhibited cytotoxic effects on the human colon adenocarcinoma DLD-1 cell line with $IC_{50}$ values of 350, 400, and 450 ppm, respectively. When T. fuciformis was extracted sequentially with ether, ethyl acetate, chloroform, and ethanol (experimental setting II), the ether extract demonstrated potent cytotoxicity with an $IC_{50}$ value of 150 ppm, followed by ethyl acetate and chloroform fractions. If the first extraction solvent was chloroform instead of ether (experimental setting III), exposure of the cell line to chloroform, ethyl acetate, and ether extracts at 1,000 ppm led to cell death. High levels of phenolic compounds were estimated for all hydrophobic extracts, which exhibited cytotoxic effects. We propose that this useful information gives additional support to our understanding of the biology and utility of this particular mushroom.

• Communications for Statistical Applications and Methods
• /
• v.27 no.3
• /
• pp.313-326
• /
• 2020

Deep Learning is the most important key to the development of Artificial Intelligence (AI). There are several distinguishable architectures of neural networks such as MLP, CNN, and RNN. Among them, we try to understand one of the main architectures called Recurrent Neural Network (RNN) that differs from other networks in handling sequential data, including time series and texts. As one of the main tasks recently in Natural Language Processing (NLP), we consider Neural Machine Translation (NMT) using RNNs. We also summarize fundamental structures of the recurrent networks, and some topics of representing natural words to reasonable numeric vectors. We organize topics to understand estimation procedures from representing input source sequences to predict target translated sequences. In addition, we apply multiple translation models with Gated Recurrent Unites (GRUs) in Keras on English-Korean sentences that contain about 26,000 pairwise sequences in total from two different corpora, colloquialism and news. We verified some crucial factors that influence the quality of training. We found that loss decreases with more recurrent dimensions and using bidirectional RNN in the encoder when dealing with short sequences. We also computed BLEU scores which are the main measures of the translation performance, and compared them with the score from Google Translate using the same test sentences. We sum up some difficulties when training a proper translation model as well as dealing with Korean language. The use of Keras in Python for overall tasks from processing raw texts to evaluating the translation model also allows us to include some useful functions and vocabulary libraries as well.

• Transactions on Electrical and Electronic Materials
• /
• v.6 no.1
• /
• pp.18-21
• /
• 2005

Monolayers of lipids on a water surface have attracted much interest as models of biological membranes, but also as precursors of multilayer systems promising many technical applications. Until now, many methodologies have been developed in order to gain a better understanding of the relationship between the structure and function of the monolayers. Maxwell displacement current (MDC) measurement has been employed to study the dielectric property of Langmuirfilms. MDC flowing across monolayers is analyzed using a rod-like molecular model. A linear relationship between the monolayer compression speed a and the molecular area Am. Compression speed a was about 30, 40, and 50 mm/min. Langmuir-Blodgett(LB) layers of Arachidic acid deposited by LB method were deposited onto slide glass as Y-type film. The structure of manufactured device is Aul Arachidic acid! AI, the number of accumulated layers are $9{\sim}21$. Also, we then examined of the Metal-Insulator-Metal(MIM) device by means of I-V. The I-V characteristics of the device are measured from -3 to+3 V. The insulation property of a thin film is better as the distance between electrodes is larger.

• Journal of the Korean Society of Systems Engineering
• /
• v.2 no.2
• /
• pp.27-32
• /
• 2006

The emerging standards since 1990's can be classified as 'system standards' (process-oriented standards) and they specify the process of an enterprise and also apply to almost all industries regardless of size, type and products. Notice that the conventional specification-oriented standards present relatively clear criteria even though the structure, performance, and terminology are defined in text-based form. However, the system standards dealing with the processes do not present a coherent guide. Therefore, it is difficult to analyze them with the same viewpoint, thereby resulting in differences in the level of understanding. This study is aimed at graphically modeling the system standards originally described in text-based form. The study has been carried out in the framework of the PMTE (Process, Methods, Tools, and Environment) paradigm. The system standard targeted here is ISO/IEC 15288. Firstly, review of the literature on the systems engineering (SE) standard/process and on the graphic model IDEF0 was done, respectively, for the parts of 'E' and 'M'. Then the SE process of the MIL-STD 499B was applied to ISO/IEC 15288 as 'P'. Finally, the graphical model was generated by AI0Wins as 'T'. As a result, the graphical model-based approach can complement the drawbacks of the text-based form.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.17 no.4
• /
• pp.63-75
• /
• 2021

It is necessary to respond to rapid technological changes such as the 4th industrial revolution and digital transformation across industries. And, a change in the university education system is necessary in a crisis situation of universities due to the rapid decrease of the school-age population. This study is aimed at practical competency with the ability to apply intelligent information technology to their job fields based on a basic understanding of intelligent information technology rather than advanced competency centered on theory and research. Instead of presenting the curriculum system diagram so that it can be flexibly applied to the design and development of the curriculum for intelligent information technology, training modules according to job and level were presented. In relation to intelligent information technology, a questionnaire was conducted for college professors, and industry experts were conducted on the derived educational modules to reflect the opinions of the industry. Industry experts said that collaboration with PBL, Capstone, and industry is necessary to improve problem-solving and communication skills.