• 한국산업정보학회논문지
• /
• 제9권3호
• /
• pp.64-70
• /
• 2004

한국어와 일본어는 동일한 어족에 속하며 비슷한 문장구조를 가지고 있어 변환중심 기계번역 방법이 효율적이다. 본 논문에서는 토큰 단위의 변환중심 한일 기계번역 시스템을 위한 변환 사전을 생성하는 방법에 관하여 기술하였다. 변환 사전이 잘 구성되면 구문분석 단계에서는 대역어를 선정하기에 적합한 정도까지의 의존트리를 생성하는 간이 파싱 만을 함으로써 필요 없는 노력을 경감시킬 수 있다. 게다가 구문해석 시에 최종의 결과 트리를 만들지 않아도 되므로 문어체 문장은 물론 입력 형태가 비정형적인 대화체 문장에서 더욱 큰 효과를 볼 수 있다. 본 논문의 변환 사전은 한국전자통신 연구원이 수집한 음성 데이터베이스로부터 추출한 말뭉치를 사용해 구성하였다. 구현한 시스템은 여행 계획영역에서 수집된 900여 발화 안의 문장을 대상으로 시험하였는데 제한된 환경에서 $92\%$, 아무런 제약이 없는 환경에서는 $81\%$의 성공률을 보였다.

• 한국통신학회논문지
• /
• 제33권8B호
• /
• pp.707-718
• /
• 2008

Traffic classification의 방법은 동적으로 변하는 application의 변화에 대처하기 위하여 페이로드나 port를 기반으로 하는 것에서 ML 알고리즘을 기반으로 하는 것으로 변하여 가고 있다. 그러나 현재의 ML 알고리즘을 이용한 traffic classification 연구는 offline 환경에 맞추어 진행되고 있다. 특히, 현재의 기존 연구들은 testing 방법으로 cross validation을 이용하여 traffic classification을 수행하고 있으며, traffic flow를 기반으로 classification 결과를 제시하고 있다. 본 논문에서는 testing방법으로 cross validation과 split validation을 이용했을 때, traffic classification의 정확도 결과를 비교한다. 또한 바이트를 기반으로 한 classification의 결과와 flow를 기반으로 한 classification의 결과를 비교해 본다. 본 논문에서는 J48, REPTree, RBFNetwork, Multilayer perceptron, BayesNet, NaiveBayes와 같은 ML 알고리즘과 다양한 feature set을 이용하여 트래픽을 분류한다. 그리고 split validation을 이용한 traffic classification에 적합한 최적의 ML 알고리즘과 feature set을 제시한다.

• 반도체디스플레이기술학회지
• /
• 제22권1호
• /
• pp.124-128
• /
• 2023

반도체 양산을 진행하며 얻어지는 여러 공정 데이터들로 사용 전압을 예측하여 에너지 효율적인 제품을 위한 목적으로 연구를 시작했다. 각각의 feature들 단독으로 전압을 예측하기 어려웠던 문제를 머신 러닝을 통해, 특히 Ensemble model을 이용함으로써 단일 모델보다 정확한 예측을 할 수 있었다. 더욱 중요한 시사점으로는 feature importance 분석을 통해 모델 예측에 영향이 큰 feature와 작은 feature에 대한 분석이다. 영향도가 높은 feature를 통해 비슷한 계열의 측정값을 늘리고, 낮은 feature 들의 문제점을 개선함으로써 차세대 제품에서 더욱 정확도 높은 모델을 위한 발판을 마련할 수 있었다.

• Computers and Concrete
• /
• 제33권1호
• /
• pp.55-75
• /
• 2024

The present study introduces an optimum performance soft computing model for predicting the compressive strength of high-performance concrete (HPC) by comparing models based on conventional (kernel-based, covariance function-based, and tree-based), advanced machine (least square support vector machine-LSSVM and minimax probability machine regressor-MPMR), and deep (artificial neural network-ANN) learning approaches using a common database for the first time. A compressive strength database, having results of 1030 concrete samples, has been compiled from the literature and preprocessed. For the purpose of training, testing, and validation of soft computing models, 803, 101, and 101 data points have been selected arbitrarily from preprocessed data points, i.e., 1005. Thirteen performance metrics, including three new metrics, i.e., a20-index, index of agreement, and index of scatter, have been implemented for each model. The performance comparison reveals that the SVM (kernel-based), ET (tree-based), MPMR (advanced), and ANN (deep) models have achieved higher performance in predicting the compressive strength of HPC. From the overall analysis of performance, accuracy, Taylor plot, accuracy metric, regression error characteristics curve, Anderson-Darling, Wilcoxon, Uncertainty, and reliability, it has been observed that model CS4 based on the ensemble tree has been recognized as an optimum performance model with higher performance, i.e., a correlation coefficient of 0.9352, root mean square error of 5.76 MPa, and mean absolute error of 4.1069 MPa. The present study also reveals that multicollinearity affects the prediction accuracy of Gaussian process regression, decision tree, multilinear regression, and adaptive boosting regressor models, novel research in compressive strength prediction of HPC. The cosine sensitivity analysis reveals that the prediction of compressive strength of HPC is highly affected by cement content, fine aggregate, coarse aggregate, and water content.

• 한국간담췌외과학회지
• /
• 제27권2호
• /
• pp.195-200
• /
• 2023

Backgrounds/Aims: We aimed to build a machine learning tool to help predict low-grade intraductal papillary mucinous neoplasms (IPMNs) in order to avoid unnecessary surgical resection. IPMNs are precursors to pancreatic cancer. Surgical resection remains the only recognized treatment for IPMNs yet carries some risks of morbidity and potential mortality. Existing clinical guidelines are imperfect in distinguishing low-risk cysts from high-risk cysts that warrant resection. Methods: We built a linear support vector machine (SVM) learning model using a prospectively maintained surgical database of patients with resected IPMNs. Input variables included 18 demographic, clinical, and imaging characteristics. The outcome variable was the presence of low-grade or high-grade IPMN based on post-operative pathology results. Data were divided into a training/validation set and a testing set at a ratio of 4:1. Receiver operating characteristics analysis was used to assess classification performance. Results: A total of 575 patients with resected IPMNs were identified. Of them, 53.4% had low-grade disease on final pathology. After classifier training and testing, a linear SVM-based model (IPMN-LEARN) was applied on the validation set. It achieved an accuracy of 77.4%, with a positive predictive value of 83%, a specificity of 72%, and a sensitivity of 83% in predicting low-grade disease in patients with IPMN. The model predicted low-grade lesions with an area under the curve of 0.82. Conclusions: A linear SVM learning model can identify low-grade IPMNs with good sensitivity and specificity. It may be used as a complement to existing guidelines to identify patients who could avoid unnecessary surgical resection.

• Advances in materials Research
• /
• 제13권1호
• /
• pp.63-86
• /
• 2024

Asphalt concrete (AC), is a mixture of bitumen and aggregates, which is very sensitive in the design of flexible pavement. In this study, the Marshall stability of the glass and carbon fiber bituminous concrete was predicted by using Artificial Neural Network (ANN), Support Vector Machine (SVM), Random Forest (RF), and M5P Tree machine learning algorithms. To predict the Marshall stability, nine inputs parameters i.e., Bitumen, Glass and Carbon fibers mixed in 100:0, 75:25, 50:50, 25:75, 0:100 percentage (designated as 100GF:0CF, 75GF:25CF, 50GF:50 CF, 25GF:75CF, 0GF:100CF), Bitumen grade (VG), Fiber length (FL), and Fiber diameter (FD) were utilized from the experimental and literary data. Seven statistical indices i.e., coefficient of correlation (CC), mean absolute error (MAE), root mean squared error (RMSE), relative absolute error (RAE), root relative squared error (RRSE), Scattering index (SI), and BIAS were applied to assess the effectiveness of the developed models. According to the performance evaluation results, Artificial neural network (ANN) was outperforming among other models with CC values as 0.9147 and 0.8648, MAE values as 1.3757 and 1.978, RMSE values as 1.843 and 2.6951, RAE values as 39.88 and 49.31, RRSE values as 40.62 and 50.50, SI values as 0.1379 and 0.2027 and BIAS value as -0.1 290 and -0.2357 in training and testing stage respectively. The Taylor diagram (testing stage) also confirmed that the ANN-based model outperforms the other models. Results of sensitivity analysis showed that the fiber length is the most influential in all nine input parameters whereas the fiber combination of 25GF:75CF was the most effective among all the fiber mixes in Marshall stability.

• 대한임베디드공학회논문지
• /
• 제9권6호
• /
• pp.307-314
• /
• 2014

Recently, software reliability and safety issues are seriously considered since failures of embedded systems may cause the damages of human lifes. For verifying and testing embedded software, execution environment including sensors and actuators should be prepared in the actual plants or virtual forms on PC. In this paper, we provide the virtual prototype based code simulation techniques and testing framework on PC. Virtual prototypes are generated by combining the Adobe's Flash SWF images corresponding to the state machine of HW or environment components. Code simulation on PC is possible by replacing the device drivers into virtual drivers which connect to virtual prototypes. Also, testing is performed by controlling the states of virtual prototype and simulators. By using these tools, embedded software can be executed in the earlier development phase and the efficiency and SW quality can be enhanced.

• 한국공작기계학회:학술대회논문집
• /
• 한국공작기계학회 2005년도 춘계학술대회 논문집
• /
• pp.71-76
• /
• 2005

The defects evaluation of the interior and the surface would be considered as vital characteristics in predicting the total life span of the steel structure. More importantly, the understandings in the interior composite of welding zone and the notifications in the presence, the formation, and the positioning of the non-metallic inclusion are necessary as well, since there were signs of relatively high defect frequency presented in the welding zone. The ultrasonic testing is a highly recommended technique chosen from among other techniques because of variety of advantages in conducting the non-destructive testing for the welding zone. However, the ultrasonic testing had technical disadvantages referred as followings; the problems due to the couplant between the PZT and the specimen, the formations that were miniature and complex, the moving subject, and the high temperature surrounding the specimen. This research was conducted to resolve the technical disadvantages of the contact ultrasonic testing by studying the non-contact ultrasonic testing where the ultrasonic waves were transferred by the laser, and revealing the specimen defects at its interior part and its surface part. The ultimate goal of this research was to develop a non-destructive evaluation applying the laser manipulated ultrasonic method for the steel structure.

• 비파괴검사학회지
• /
• 제32권6호
• /
• pp.686-695
• /
• 2012

This paper proposes a noncontact thermography technique for fatigue crack evaluation under a cyclic tensile loading. The proposed technique identifies and localizes an invisible fatigue crack without scanning, thus making it possible to instantaneously evaluate an incipient fatigue crack. Based on a thermoelastic theory, a new fatigue crack evaluation algorithm is proposed for the fatigue crack-tip localization. The performance of the proposed algorithm is experimentally validated. To achieve this, the cyclic tensile loading is applied to a dog-bone shape aluminum specimen using a universal testing machine, and the corresponding thermal responses induced by thermoelastic effects are captured by an infrared camera. The test results confirm that the fatigue crack is well identified and localized by comparing with its microscopic images.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 춘계학술대회논문집
• /
• pp.969-973
• /
• 2007

The heavy-weight floor impact sound field of the receiving room in a testing building with bearing wall structure was investigated using bang machine and impact ball. The sound field was investigated through the impact sound pressure level distribution by the field measurement and computational analysis. Predicted sound field using the computational analysis agree with measurement result in the low frequency band. Result shows that standard deviations of the single number rating value are about 2dB in each impact source. Particularly, impact sound pressure level at 120cm height in 63Hz octave band was 5dB lower than spatial averaging value. It was found that receiving positions in the ministry of construction and transportation notice should be reconsidered.