• Transactions of Materials Processing
• /
• v.29 no.5
• /
• pp.282-289
• /
• 2020

Flexible roll forming is an advanced sheet-metal-forming process that allows the production of parts with various cross-sections. During the flexible process, material is subjected to three-dimensional deformation such as transverse bending, inhomogeneous elongations, or contraction. Because of the effects of process variables on the quality of the roll-formed products, the approaches used to investigate the roll-forming process have been largely dependent on experience and trial- and-error methods. Web-warping is one of the major shape defects encountered in flexible roll forming. In this study, an SVR model was developed to predict the web-warping during the flexible roll forming process. In the development of the SVR model, three process parameters, namely the forming-roll speed condition, leveling-roll height, and bend angle were considered as the model inputs, and the web-warping height was used as the response variable for three blank shapes; rectangular, concave, and convex shape. MATLAB software was used to train the SVR model and optimize three hyperparameters (λ, ε, and γ). To evaluate the SVR model performance, the statistical analysis was carried out based on the three indicators: the root-mean-square error, mean absolute error, and relative root-mean-square error.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.3
• /
• pp.1089-1098
• /
• 2018

This study aims at developing and applying a hybrid model to the wind power prediction (WPP). The hybrid model for a very-short-term WPP (VSTWPP) is achieved through analytical data, multiple linear regressions and least square methods (MLR&LS). The data used in our hybrid model are based on the historical records of wind power from an offshore region. In this model, the WPP is achieved in four steps: 1) transforming historical data into ratios; 2) predicting the wind power using the ratios; 3) predicting rectification ratios by the total wind power; 4) predicting the wind power using the proposed rectification method. The proposed method includes one-step and multi-step predictions. The WPP is tested by applying different models, such as the autoregressive moving average (ARMA), support vector machine (SVM), and artificial neural network (ANN). The results of all these models confirmed the validity of the proposed hybrid model in terms of error as well as its effectiveness. Furthermore, forecasting errors are compared to depict a highly variable WPP, and the correlations between the actual and predicted wind powers are shown. Simulations are carried out to definitely prove the feasibility and excellent performance of the proposed method for the VSTWPP versus that of the SVM, ANN and ARMA models.

• ETRI Journal
• /
• v.44 no.3
• /
• pp.462-475
• /
• 2022

Although researchers have proposed numerous techniques for speech emotion recognition, its performance remains unsatisfactory in many application scenarios. In this study, we propose a speech emotion recognition model based on a genetic algorithm (GA)-decision tree (DT) fusion of deep and acoustic features. To more comprehensively express speech emotional information, first, frame-level deep and acoustic features are extracted from a speech signal. Next, five kinds of statistic variables of these features are calculated to obtain utterance-level features. The Fisher feature selection criterion is employed to select high-performance features, removing redundant information. In the feature fusion stage, the GA is is used to adaptively search for the best feature fusion weight. Finally, using the fused feature, the proposed speech emotion recognition model based on a DT support vector machine model is realized. Experimental results on the Berlin speech emotion database and the Chinese emotion speech database indicate that the proposed model outperforms an average weight fusion method.

• Geomechanics and Engineering
• /
• v.29 no.3
• /
• pp.249-258
• /
• 2022

A disc cutter is an excavation tool on a tunnel boring machine (TBM) cutterhead; it crushes and cuts rock mass while the machine excavates using the cutterhead's rotational movement. Disc cutter wear occurs naturally. Thus, along with the management of downtime and excavation efficiency, abrasioned disc cutters need to be replaced at the proper time; otherwise, the construction period could be delayed and the cost could increase. The most common prediction models for TBM performance and for the disc cutter lifetime have been proposed by the Colorado School of Mines and Norwegian University of Science and Technology. However, design parameters of existing models do not well correspond to the field values when a TBM encounters complex and difficult ground conditions in the field. Thus, this study proposes a series of machine learning models to predict the disc cutter lifetime of a shield TBM using the excavation (machine) data during operation which is response to the rock mass. This study utilizes five different machine learning techniques: four types of classification models (i.e., K-Nearest Neighbors (KNN), Support Vector Machine, Decision Tree, and Staking Ensemble Model) and one artificial neural network (ANN) model. The KNN model was found to be the best model among the four classification models, affording the highest recall of 81%. The ANN model also predicted the wear rate of disc cutters reasonably well.

• Nuclear Engineering and Technology
• /
• v.53 no.12
• /
• pp.4022-4032
• /
• 2021

In this study, machine learning (ML) techniques were used to model surveillance test data of nuclear power plants from an international database of the ASTM E10.02 committee. Regression modeling was conducted using various techniques, including Cubist, XGBoost, and a support vector machine. The root mean square deviation of each ML model for the baseline dataset was less than that of the ASTM E900-15 nonlinear regression model. With respect to the interpolation, the ML methods provided excellent predictions with relatively few computations when applied to the given data range. The effect of the explanatory variables on the transition temperature shift (TTS) for the ML methods was analyzed, and the trends were slightly different from those for the ASTM E900-15 model. ML methods showed some weakness in the extrapolation of the fluence in comparison to the ASTM E900-15, while the Cubist method achieved an extrapolation to a certain extent. To achieve a more reliable prediction of the TTS, it was confirmed that advanced techniques should be considered for extrapolation when applying ML modeling.

• Current Optics and Photonics
• /
• v.6 no.1
• /
• pp.92-103
• /
• 2022

The development of midinfrared (mid-IR) quantum cascade lasers (QCLs) has enabled rapid high-contrast measurement of the mid-IR spectra of biological tissues. Several studies have compared the differences between the mid-IR spectra of colon cancer and noncancerous colon tissues. Most mid-IR spectrum classification studies have been proposed as machine-learning-based algorithms, but this results in deviations depending on the initial data and threshold values. We aim to develop a process for classifying colon cancer and noncancerous colon tissues through a deep-learning-based convolutional-neural-network (CNN) model. First, we image the midinfrared spectrum for the CNN model, an image-based deep-learning (DL) algorithm. Then, it is trained with the CNN algorithm and the classification ratio is evaluated using the test data. When the tissue microarray (TMA) and routine pathological slide are tested, the ML-based support-vector-machine (SVM) model produces biased results, whereas we confirm that the CNN model classifies colon cancer and noncancerous colon tissues. These results demonstrate that the CNN model using midinfrared-spectrum images is effective at classifying colon cancer tissue and noncancerous colon tissue, and not only submillimeter-sized TMA but also routine colon cancer tissue samples a few tens of millimeters in size.

• Geomechanics and Engineering
• /
• v.28 no.1
• /
• pp.65-75
• /
• 2022

Time and cost of construction are key factors in decision-making during a tunnel project's planning and design phase. Estimations of time and cost of tunnel construction projects are subject to significant uncertainties caused by uncertain geotechnical and geological conditions. The Gaussian Process Regression (GPR) technique for predicting ground condition and construction time and cost of mountain tunnel projects is used in this work. The GPR model is trained with data from past mountain tunnel projects. The model is applied to a case study in which the predicted time and cost of tunnel construction using the GPR model are compared with the actual construction time and cost for model validation and reducing the uncertainty for the future projects. In addition, the results obtained from the GPR have been compared with to other models of artificial neural network (ANN) and support vector regression (SVR) that the GPR model provides more accurate results.

• Journal of KIISE:Software and Applications
• /
• v.36 no.9
• /
• pp.691-696
• /
• 2009

Introduction: There are two types of asthma according to aspirin hypersensitivity: aspirin intolerant asthma (AIA) and aspirin tolerant asthma (ATA). The genetic risk factors that are related with asthma have been investigated intensively and extensively. However the combinatory effects of single nucleotide polymorphisms (SNPs) have hardly been evaluated. In this paper we searched the best set of SNPs that are useful to diagnose the two types of asthma. Methods: We examined 246 asthmatic patients (94 having aspirin intolerant asthma and 152 having aspirin tolerant asthma) and analyzed 25 SNPs typed in them, which are suspected to be associated with asthma. Normalized mutual information values of combinations of typed SNPs are calculated, and those with high normalized mutual information values are selected. We use support vector machines to evaluate the prediction accuracy of the selected combinations. Results: The best combination model turns out four-locus and consists of ALOX5_p1_1708, B2ADR_q1_46, CCR3_p1_520, CysLTR1_p1_634. Its normalized mutual information value is 0.053 and the accuracy in predicting ATA disease risk among asthmatic patients is 71.14%.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.555-562
• /
• 2016

Semantic Role Labeling (SRL) is to determine the semantic relation of a predicate and its argu-ments in a sentence. But Korean semantic role labeling has faced on difficulty due to its different language structure compared to English, which makes it very hard to use appropriate approaches developed so far. That means that methods proposed so far could not show a satisfied perfor-mance, compared to English and Chinese. To complement these problems, we focus on suffix information analysis, such as josa (case suffix) and eomi (verbal ending) analysis. Korean lan-guage is one of the agglutinative languages, such as Japanese, which have well defined suffix structure in their words. The agglutinative languages could have free word order due to its de-veloped suffix structure. Also arguments with a single morpheme are then labeled with statistics. In addition, machine learning algorithms such as Support Vector Machine (SVM) and Condi-tional Random Fields (CRF) are used to model SRL problem on arguments that are not labeled at the suffix analysis phase. The proposed method is intended to reduce the range of argument instances to which machine learning approaches should be applied, resulting in uncertain and inaccurate role labeling. In experiments, we use 15,224 arguments and we are able to obtain approximately 83.24% f1-score, increased about 4.85% points compared to the state-of-the-art Korean SRL research.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.1-9
• /
• 2021

In this paper, we propose a training algorithm of support vector machine (SVM) with a sensitive variable. Although machine learning models enable automatic decision making in the real world applications, regulations prohibit sensitive information from being used to protect privacy. In particular, the privacy protection of the legally protected attributes such as race, gender, and disability is compulsory. We present an efficient least square SVM (LSSVM) training algorithm using a fully homomorphic encryption (FHE) to protect a partial sensitive attribute. Our framework posits that data owner has both non-sensitive attributes and a sensitive attribute while machine learning service provider (MLSP) can get non-sensitive attributes and an encrypted sensitive attribute. As a result, data owner can obtain the encrypted model parameters without exposing their sensitive information to MLSP. In the inference phase, both non-sensitive attributes and a sensitive attribute are encrypted, and all computations should be conducted on encrypted domain. Through the experiments on real data, we identify that our proposed method enables to implement privacy-preserving sensitive LSSVM with FHE that has comparable performance with the original LSSVM algorithm. In addition, we demonstrate that the efficient sensitive LSSVM with FHE significantly improves the computational cost with a small degradation of performance.