• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.3
• /
• pp.399-407
• /
• 2019

In Korea, the construction industry has been known to have the highest risk of safety accidents compared to other industries. Therefore, in order to improve safety in the construction industry, several researches have been carried out from the past. This study aims at improving safety of labors in construction site by constructing an effective automatic safety helmet detection system using object detection algorithm based on image data of construction field. Deep learning was conducted using Region-based Fully Convolutional Network (R-FCN) which is one of the object detection algorithms based on Convolutional Neural Network (CNN) with Transfer Learning technique. Learning was conducted with 1089 images including human and safety helmet collected from ImageNet and the mean Average Precision (mAP) of the human and the safety helmet was measured as 0.86 and 0.83, respectively.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.711-714
• /
• 2024

In the dynamic landscape of modern machine learning, Federated Learning (FL) has emerged as a compelling paradigm designed to enhance privacy by enabling participants to collaboratively train models without sharing their private data. Specifically, Distillation-based Federated Learning, like Federated Learning with Model Distillation (FedMD), Federated Gradient Encryption and Model Sharing (FedGEMS), and Differentially Secure Federated Learning (DS-FL), has arisen as a novel approach aimed at addressing Non-IID data challenges by leveraging Federated Learning. These methods refine the standard FL framework by distilling insights from public dataset predictions, securing data transmissions through gradient encryption, and applying differential privacy to mask individual contributions. Despite these innovations, our survey identifies persistent vulnerabilities, particularly concerning the susceptibility to logit inversion attacks where malicious actors could reconstruct private data from shared public predictions. This exploration reveals that even advanced Distillation-based Federated Learning systems harbor significant privacy risks, challenging the prevailing assumptions about their security and underscoring the need for continued advancements in secure Federated Learning methodologies.

• The Journal of Information Systems
• /
• v.20 no.4
• /
• pp.181-204
• /
• 2011

Recently, the development of ICT(information & communications technology) with the advent of new media paradigm shift in learning has brought a dramatic impact on the competitiveness of universities. The previous studies on the academic performance of e-learning mainly targeted on e-learning users, studying additional synchronization and system quality factors to measure academic performance. This study empirically analyzed the learning flow and academic performance considering both DeLone & McLean model system quality and synchronizing factors based on ARCS model. Relating to quality and synchronization factors, the academic performance of e-learning system was tested, and the difference between learning flow and academic performance was also analyzed based on time-series data, by the test difference(in the beginning, during, and final of the semester). The results of the study are as follows. First, the study shows that both system quality and synchronization directly affected the learning performance. Thus, when designing e-learning system, it is necessary to consider these two factors at the same time. Second, the indirectly mediating effect on the system quality and synchronization factors turned out to be significant in learning flow. Third, the result of regression analysis on the contents of utilizing dummy variable presents that the teacher's explanation has greater influence than multimedia has to the academic performance, and furthermore, the test difference showed no significance. Further research should be undertaken to consider the learner's degree of acceptance which reflects various aspects for building m-learning or u-learning.

• Journal of The Korean Astronomical Society
• /
• v.52 no.6
• /
• pp.217-225
• /
• 2019

We apply a modified Convolutional Neural Network (CNN) model in conjunction with transfer learning to predict whether an active region (AR) would produce a ≥C-class or ≥M-class flare within the next 24 hours. We collect line-of-sight magnetogram samples of ARs provided by the SHARP from May 2010 to September 2018, which is a new data product from the HMI onboard the SDO. Based on these AR samples, we adopt the approach of shuffle-and-split cross-validation (CV) to build a database that includes 10 separate data sets. Each of the 10 data sets is segregated by NOAA AR number into a training and a testing data set. After training, validating, and testing our model, we compare the results with previous studies using predictive performance metrics, with a focus on the true skill statistic (TSS). The main results from this study are summarized as follows. First, to the best of our knowledge, this is the first time that the CNN model with transfer learning is used in solar physics to make binary class predictions for both ≥C-class and ≥M-class flares, without manually engineered features extracted from the observational data. Second, our model achieves relatively high scores of TSS = 0.640±0.075 and TSS = 0.526±0.052 for ≥M-class prediction and ≥C-class prediction, respectively, which is comparable to that of previous models. Third, our model also obtains quite good scores in five other metrics for both ≥C-class and ≥M-class flare prediction. Our results demonstrate that our modified CNN model with transfer learning is an effective method for flare forecasting with reasonable prediction performance.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.731-739
• /
• 1998

The probabilistic relaxation(PR) scheme based on the conditional probability and probability space partition has the important property that when its compatibility coefficient matrix (CCM) has uniform components it can classify m-dimensional probabilistic distribution vectors into different classes. When consistency or inconsistency measures have been defined, the properties of PRs are completely determined by the compatibility coefficients among labels of labeled objects and influence weight among labeled objects. In this paper we study the properties of PR in which both compatibility coefficients and influence weights are uniform, and then a learning rule for such PR system is derived. Experiments have been performed to verify the effectiveness of the learning rule.

• Journal of Korea Multimedia Society
• /
• v.16 no.11
• /
• pp.1338-1347
• /
• 2013

Face recognition is a science of automatically identifying individuals based their unique facial features. In order to avoid overfitting and reduce the computational reduce the computational burden, a new face recognition algorithm using PCA-fisher linear discriminant (PCA-FLD) and fuzzy radial basis function neural network (RBFNN) is proposed in this paper. First, face features are extracted by the principal component analysis (PCA) method. Then, the extracted features are further processed by the Fisher's linear discriminant technique to acquire lower-dimensional discriminant patterns, the processed features will be considered as the input of the fuzzy RBFNN. As a widely applied algorithm in fuzzy RBF neural network, BP learning algorithm has the low rate of convergence, therefore, an improved learning algorithm based on Levenberg-Marquart (L-M) for fuzzy RBF neural network is introduced in this paper, which combined the Gradient Descent algorithm with the Gauss-Newton algorithm. Experimental results on the ORL face database demonstrate that the proposed algorithm has satisfactory performance and high recognition rate.

• Research in Mathematical Education
• /
• v.9 no.3 s.23
• /
• pp.269-273
• /
• 2005

Nowadays application of modem achievements of information technologies in science, engineering and education is usual phenomenon. Application of these technologies allows easily creating new methods of learning of mathematics. More of new methods of creation of multimedia electronic manuals on high mathematics are founded to application of multimedia and communication opportunities of the computer. But application only multimedia and communication opportunities of the computer at creation of multimedia electronic manuals on high mathematics is insufficient to elimination of 'gap' between training and studying high mathematics. So, we offer a new way of the decision of this problem: creation of a multimedia electronic manual on high mathematics with built-in a mathematical environment MathCad in the national language.

• Journal of the Korean Institute of Rural Architecture
• /
• v.14 no.1
• /
• pp.21-28
• /
• 2012

Campus facilities were recently remodeled to provide the substantial learning rights of handicapped students in many campus to embody the dignity and value as man. So this study aims to identify the exterior connection facilities for handicapped students of S and D campuses. The summaries of this research are as follows. Installations of even crossing area(1.5mx1.5m) per 50m and even rest area(1.5mx1.5m) per 30m in walking or access ramp. Improving in accordance with exterior connection facility repairing master plan in S campus. Bringing down an angle degrees of the inclined walking or access ramp in D campus. Installation of exterior braille guide sign for blind students. All handicapped students must be guaranteed the same learning rights as normal men to remove obstacles as the upper mentioned imperfections in using exterior campus facilities.

• Research in Mathematical Education
• /
• v.17 no.4
• /
• pp.221-241
• /
• 2013

The data for the present paper was a part of a large research project conducted to assess preservice teachers' knowledge related to fractions and place value at a southwestern public university in 2007. The study utilized convenience sampling, consisting of 150 elementary preservice teachers who were enrolled in a mathematics methods course before their student teaching. The results demonstrated preservice teachers' knowledge of teaching comparison, addition, subtraction, and multiplication of fractions was insufficient even though these should be basic knowledge. Teacher preparation programs should emphasize profound knowledge for teaching fractions using representations.

• ETRI Journal
• /
• v.44 no.3
• /
• pp.438-449
• /
• 2022

This study proposes a dependency grammar-based self-attention multilayered bidirectional long short-term memory (DG-M-Bi-LSTM) model for subject-predicate-object (SPO) tuple recognition from natural language (NL) sentences. To add recent knowledge to the knowledge base autonomously, it is essential to extract knowledge from numerous NL data. Therefore, this study proposes a high-accuracy SPO tuple recognition model that requires a small amount of learning data to extract knowledge from NL sentences. The accuracy of SPO tuple recognition using DG-M-Bi-LSTM is compared with that using NL-based self-attention multilayered bidirectional LSTM, DG-based bidirectional encoder representations from transformers (BERT), and NL-based BERT to evaluate its effectiveness. The DG-M-Bi-LSTM model achieves the best results in terms of recognition accuracy for extracting SPO tuples from NL sentences even if it has fewer deep neural network (DNN) parameters than BERT. In particular, its accuracy is better than that of BERT when the learning data are limited. Additionally, its pretrained DNN parameters can be applied to other domains because it learns the structural relations in NL sentences.