• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.11
• /
• pp.1634-1640
• /
• 2017

The study aims to distinguish hemiplegic gait and normal gait using simple wearable device and classification algorithm. Thus, we developed a wearable system equipped three axis accelerometer and three axis gyroscope. The developed wearable system was verified by clinical experiment. In experiment, twenty one normal subjects and twenty one patients undergoing stroke treatment were participated. Based on the measured inertial signal, a random forest algorithm was used to classify hemiplegic gait. Four-fold cross validation was applied to ensure the reliability of the results. To select optimal attributes, we applied the forward search algorithm with 10 times of repetition, then selected five most frequently attributes were chosen as a final attribute. The results of this study showed that 95.2% of accuracy in hemiplegic gait and normal gait classification and 77.4% of accuracy in hemiplegic-side and normal gait classification.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.83-86
• /
• 2000

본 논문에서는 비정형 단위를 사용한 음성 합성 시스템의 합성음에 대한 자연성을 향상시키기 위한 휴지 구간 추출 및 휴지 지속시간 예측 모델을 제안한다. 제안된 휴지 지속시간 예측 모델은 트리 기반 모델링 기법 중 하나인 CART (Classification And Regression Trees)방법을 이용하였다. 이를 위해 남성 단일 화자가 발성한 6,220개의 어절경계 포함하는 총 400문장의 문 음성 데이터베이스를 구축하였고, 이 데이터베이스로부터 V-fold Cross-Validation 방법에 의해 최적의 트리를 결정하였다. 이 모델을 평가한 결과, 휴지 구간 추출 정확율은 $81\%$로 휴지 구간 존재 추출 정확율은 $83\%, 휴지 구간 비존재 추출 정확율은 $80\%이었고, 실 휴지지속시간과 예측 휴지지속시간과의 다중상관 계수는 0.84로, 오차 범위 20ms 이내에서 의 정 확율은 $88\%$ 이었다. 또한, 휴지지속시간을 예측하여 적용한 합성음을 청취 실험한 결과 자연 음성과 대체적으로 유사하게 나타났다.

• BMB Reports
• /
• v.43 no.10
• /
• pp.670-676
• /
• 2010

In this paper, a novel approach, ELM-PCA, is introduced for the first time to predict protein subcellular localization. Firstly, Protein Samples are represented by the pseudo amino acid composition (PseAAC). Secondly, the principal component analysis (PCA) is employed to extract essential features. Finally, the Elman Recurrent Neural Network (RNN) is used as a classifier to identify the protein sequences. The results demonstrate that the proposed approach is effective and practical.

• Journal of Distribution Science
• /
• v.11 no.3
• /
• pp.5-11
• /
• 2013

Purpose - Going concern is one of fundamental concepts in accounting and auditing and sometimes the assessment of a company's going concern status that is a tough process. Various going concern prediction models' based on statistical and data mining methods help auditors and stakeholders suggested in the previous literature. Research design - This paper employs a data mining approach to prediction of going concern status of Iranian firms listed in Tehran Stock Exchange using Particle Swarm Optimization. To reach this goal, at the first step, we used the stepwise discriminant analysis it is selected the final variables from among of 42 variables and in the second stage; we applied a grid-search technique using 10-fold cross-validation to find out the optimal model. Results - The empirical tests show that the particle swarm optimization (PSO) model reached 99.92% and 99.28% accuracy rates for training and holdout data. Conclusions - The authors conclude that PSO model is applicable for prediction going concern of Iranian listed companies.

• Journal of Computing Science and Engineering
• /
• v.11 no.1
• /
• pp.24-31
• /
• 2017

Face-based video retrieval has become an active and important branch of intelligent video analysis. Face profiling and matching is a fundamental step and is crucial to the effectiveness of video retrieval. Although many algorithms have been developed for processing static face images, their effectiveness in face-based video retrieval is still unknown, simply because videos have different resolutions, faces vary in scale, and different lighting conditions and angles are used. In this paper, we combined content-based and semantic-based image analysis techniques, and systematically evaluated four mainstream local features to represent face images in the video retrieval task: Harris operators, SIFT and SURF descriptors, and eigenfaces. Results of ten independent runs of 10-fold cross-validation on datasets consisting of TED (Technology Entertainment Design) talk videos showed the effectiveness of our approach, where the SIFT descriptors achieved an average F-score of 0.725 in video retrieval and thus were the most effective, while the SURF descriptors were computed in 0.3 seconds per image on average and were the most efficient in most cases.

• Nuclear Engineering and Technology
• /
• v.56 no.1
• /
• pp.100-105
• /
• 2024

In this study, we propose an alternative approach using Artificial Neural Networks (ANN) for determining Mass Attenuation Coefficients (MAC) in various glass systems. This method takes into account the weights of glass compositions, density, and photon energy as input features. The ANN model was trained and tested on a dataset consisting of 650 data points and subsequently validated through a K-fold cross-validation procedure. Our findings demonstrate a high level of accuracy, with R² values ranging from 0.90 to 0.99. Additionally, the model exhibits robust extrapolation capabilities with an R² score of 0.87 for predicting MAC values in a new glass system. Furthermore, this approach significantly reduces the need for costly and time-consuming computations and experiments, making it a potential tool for selecting materials for effective radiation protection.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.132-132
• /
• 2021

Access to accurate spatial precipitation in many hydrological studies is necessary. Existence of many mountains with diverse topography in South Korea causes different spatial distribution of precipitation. Rain gauge stations show accurate precipitation information in points, but due to the limited use of rain gauge stations and the difficulty of accessing them, there is not enough accurate information in the whole area. Weather radars can provide an integrated precipitation information spatially. Despite this, weather radar data have some errors that can not provide accurate data, especially in heavy rainfall. In this study, some location-based variable like aspect, elevation, plan curvature, profile curvature, slope and distance from the sea which has most effect on rainfall was considered. Then Automatic Weather Station data was used for spatial training of variables in each event. According to this, K-fold cross-validation method was combined with Adaptive Neuro-Fuzzy Inference System. Based on this, 80% of Automatic Weather Station data was used for training and validation of model and 20% was used for testing and evaluation of model. Finally, spatial distribution of precipitation for 1×1 km resolution in Gwangdeoksan radar station was estimates. The results showed a significant decrease in RMSE and an increase in correlation with the observed amount of precipitation.

• Geomechanics and Engineering
• /
• v.31 no.3
• /
• pp.265-279
• /
• 2022

Tunnel convergence prediction is essential for the safe construction and design of tunnels. This study proposes five machine learning models of deep neural network (DNN), K-nearest neighbors (KNN), Gaussian process regression (GPR), support vector regression (SVR), and decision trees (DT) to predict the convergence phenomenon during or shortly after the excavation of tunnels. In this respect, a database including 650 datasets (440 for training, 110 for validation, and 100 for test) was gathered from the previously constructed tunnels. In the database, 12 effective parameters on the tunnel convergence and a target of tunnel wall convergence were considered. Both 5-fold and hold-out cross validation methods were used to analyze the predicted outcomes in the ML models. Finally, the DNN method was proposed as the most robust model. Also, to assess each parameter's contribution to the prediction problem, the backward selection method was used. The results showed that the highest and lowest impact parameters for tunnel convergence are tunnel depth and tunnel width, respectively.

• Journal of Microbiology and Biotechnology
• /
• v.26 no.1
• /
• pp.66-71
• /
• 2016

PikD is a widely known pathway-specific regulator for controlling pikromycin production in Streptomyces venezuelae ATCC 15439, which is a representative of the large ATP-binding regulator of the LuxR family (LAL) in Streptomyces sp. RapH and FkbN also belong to the LAL family of transcriptional regulators, which show greatest homology with the ATP-binding motif and helix-turn-helix DNA-binding motif of PikD. Overexpression of pikD and heterologous expression of rapH and fkbN led to enhanced production of pikromycin by approximately 1.8-, 1.6-, and 1.6-fold in S. venezuelae, respectively. Cross-complementation of rapH and fkbN in the pikD deletion mutant (ΔpikD) restored pikromycin and derived macrolactone production. Overall, these results show that heterologous expression of rapH and fkbN leads to the overproduction of pikromycin and its congeners from the pikromycin biosynthetic pathway in S. venezuelae, and they have the same functionality as the pathwayspecific transcriptional activator for the pikromycin biosynthetic pathway in the ΔpikD strain. These results also show extensive "cross-communication" between pathway-specific regulators of streptomycetes and suggest revision of the current paradigm for pathwayspecific versus global regulation of secondary metabolism in Streptomyces species.

• Resources Recycling
• /
• v.26 no.2
• /
• pp.46-55
• /
• 2017

In this study, the design methodology of Radial Basis Function Neural Networks is developed with the aid of Laser Induced Breakdown Spectroscopy and also applied to the practical plastics sorting system. To identify black plastics such as ABS, PP, and PS, RBFNNs classifier as a kind of intelligent algorithms is designed. The dimensionality of the obtained input variables are reduced by using PCA and divided into several groups by using K-means clustering which is a kind of clustering techniques. The entire data is split into training data and test data according to the ratio of 4:1. The 5-fold cross validation method is used to evaluate the performance as well as reliability of the proposed classifier. In case of input variables and clusters equal to 5 respectively, the classification performance of the proposed classifier is obtained as 96.78%. Also, the proposed classifier showed superiority in the viewpoint of classification performance where compared to other classifiers.