• Journal of Information Processing Systems
• /
• 제18권4호
• /
• pp.535-548
• /
• 2022

Using a rich resource language to classify sentiments in a language with few resources is a popular subject of research in natural language processing. Burmese is a low-resource language. In light of the scarcity of labeled training data for sentiment classification in Burmese, in this study, we propose a method of transfer learning for sentiment analysis of a language that uses the feature transfer technique on sentiments in English. This method generates a cross-language word-embedding representation of Burmese vocabulary to map Burmese text to the semantic space of English text. A model to classify sentiments in English is then pre-trained using a convolutional neural network and an attention mechanism, where the network shares the model for sentiment analysis of English. The parameters of the network layer are used to learn the cross-language features of the sentiments, which are then transferred to the model to classify sentiments in Burmese. Finally, the model was tuned using the labeled Burmese data. The results of the experiments show that the proposed method can significantly improve the classification of sentiments in Burmese compared to a model trained using only a Burmese corpus.

• Geomechanics and Engineering
• /
• 제29권5호
• /
• pp.487-496
• /
• 2022

Most of the pile's vertical static load tests in construction sites are the proof load tests, which is difficult to accurately estimate the ultimate bearing capacity and analyze the reliability of piles. Therefore, a reliability analysis method based on the proof load-settlement (Q-s) data is proposed in this study. In this proposed method, a simple ultimate limit state function based on the hyperbolic model is established, where the random variables of reliability analysis include the model factor of the ultimate bearing capacity and the fitting parameters of the hyperbolic model. The model factor M = R_uR / R_uP is calculated based on the available destructive Q-s data, where the real value of the ultimate bearing capacity (R_uR) is obtained by the complete destructive Q-s data; the predicted value of the ultimate bearing capacity (R_uP) is obtained by the proof Q-s data, a part of the available destructive Q-s data, that before the predetermined load determined by the pile test report. The results demonstrate that the proposed method can easy and effectively perform the reliability analysis based on the proof Q-s data.

• Geomechanics and Engineering
• /
• 제38권1호
• /
• pp.45-56
• /
• 2024

The weakening effect of water on rocks is one of the main factors inducing deformation and failure in rock engineering. To clarify this weakening effect, immersion tests and post-immersion triaxial compression tests were conducted on sandstone and mudstone. The results showed that the strength of water-immersed sandstone decreases with increasing immersion time, exhibiting an exponential relationship. Similarly, the strength of water-immersed mudstone decreases with increasing environmental humidity, also following an exponential relationship. Subsequently, a statistical damage model for water-weakened rocks was proposed, changes in elastic modulus to describe the weakening effect of water. The model effectively simulated the stress-strain relationships of water-affected sandstone and mudstone under compression. The R² values between the theoretical and experimental peak values ranged from 0.962 to 0.996, and the MAPE values fell between 3.589% and 9.166%, demonstrating the model's effectiveness and reliability. The damage process of water-saturated rocks corresponds to five stages: compaction stage - no damage, elastic stage - minor damage, crack development stage - rapid damage increase, post-peak residual stage - continuous damage increase, and sliding stage - damage completion. This study provides a foundational reference for researching the fracture characteristics of overlying strata during coal mining under complex hydrogeological conditions.

• Steel and Composite Structures
• /
• 제52권4호
• /
• pp.475-485
• /
• 2024

The mechanical behavior of the steel tube encased high-strength concrete (STHC) composite walls under constant axial load and cyclically increasing lateral load was studied. Conclusions are drawn based on experimental observations, grey evolutionary algorithm and finite element (FE) simulations. The use of steel tube wall panels improved the load capacity and ductility of the specimens. STHC composite walls withstand more load cycles and show more stable hysteresis performance than conventional high strength concrete (HSC) walls. After the maximum load, the bearing capacity of the STHC composite wall was gradually reduced, and the wall did not collapse under the influence of the steel pipe. For analysis of the bending capacity of STHC composite walls based on artificial intelligence tools, an analysis model is proposed that takes into account the limiting effect of steel pipes. The results of this model agree well with the test results, indicating that the model can be used to predict the bearing capacity of STHC composite walls. Based on a reasonable material constitutive model and the limiting effect of steel pipes, a finite element model of the STHC composite wall was created. The finite elements agree well with the experimental results in terms of hysteresis curve, load-deformation curve and peak load.

• Journal of Ship and Ocean Technology
• /
• 제5권2호
• /
• pp.27-48
• /
• 2001

Turbulent flow calculations are performed for the two modern practical VLCCs with the sable forebody and the slightly different afterbody, i.e. KVLCC and KVLCC2. Three $\textsc{k}-\varepsilon$ turbulence models are tested to investigate the differences caused by the turbulence models. The calculated results around the two VLCC hull forms using O-O grid topology and profile-fitted surface meshes are compared to the measured data from towing tank experiment. The realizable $\textsc{k}-\varepsilon$model provided realistic wake distribution with hook-like shape, while the standard and RNG-based $\textsc{k}-\varepsilon$models failed. It is very encouraging to see that the CFD with relatively simple turbulence closure can tell the difference quantitatively as well as qualitatively for the two hull forms with stern frameline modification.

• Structural Engineering and Mechanics
• /
• 제16권6호
• /
• pp.643-654
• /
• 2003

Two real-time modeling prediction (RMP) schemes are presented in this paper for analyzing the behavior of deep excavations during construction. The first RMP scheme is developed from the traditional AR(p) model. The second is based on the simplified Elman-style recurrent neural networks. An on-line learning algorithm is introduced to describe the dynamic behavior of deep excavations. As a case study, in-situ measurements of an excavation were recorded and the measured data were used to verify the reliability of the two schemes. They proved to be both effective and convenient for predicting the behavior of deep excavations during construction. It is shown through the case study that the RMP scheme based on the neural network is more accurate than that based on the traditional AR(p) model.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1211-1215
• /
• 2003

Accurate mathematical models of spacecraft components are an essential of spacecraft attitude control system design, analysis and simulation. Gyro is one of the most important spacecraft components used for attitude propagation and control. Gyro errors may seriously degrade the accuracy of the calculated spacecraft angular rate and of attitude estimates due to inherent drift and bias errors. In order to validate this model, nominal case simulation has been performed and compared for the low range mode and high range mode, respectively. In this paper, a mathematical model of gyro containing the relationships for predicting spacecraft angular rate and disturbances is proposed.

• 한국자동차공학회논문집
• /
• 제7권9호
• /
• pp.48-55
• /
• 1999

In this paper, a zero-dimensional two zone model is developed to investigate the effects of variable valve timing on combustion process in SI engine. The simulation results show that the predicted data has good agreement with experimental ones. The useful information of combustion process such like residual gas fraction cylinder pressure, cylinder temperature and NO concentration can be obtained and the effects of engine variables on combustion processes and performances can be evaluated.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2016년도 하계학술대회
• /
• pp.264-266
• /
• 2016

During the $2^{nd}$ JVET (Joint Group on Future Video Coding Technology Exploration) meeting, up to 22 coding tools focusing on Future Video Coding (FVC) were proposed. Despite that the application of proposed coding tools has a considerable performance enhancement, however, the encoding time of Joint Exploration Model (JEM) software is over 20 times for All Intra coding mode, 6 times for Random Access coding mode, of HEVC reference model (HM), and decoding time is 1.6 times for All Intra coding mode, 7.9 times for Random Access coding mode, of HM. This paper focuses on analyzing the complexity of the JEM software compared with HM.

• Communications for Statistical Applications and Methods
• /
• 제26권6호
• /
• pp.635-644
• /
• 2019

Class probability is a fundamental target in classification that contains complete classification information. In this article, we propose a class probability estimation method when the predictor is functional. Motivated by Wang et al. (Biometrika, 95, 149-167, 2007), our estimator is obtained by training a sequence of functional weighted support vector machines (FWSVM) with different weights, which can be justified by the Fisher consistency of the hinge loss. The proposed method can be extended to multiclass classification via pairwise coupling proposed by Wu et al. (Journal of Machine Learning Research, 5, 975-1005, 2004). The use of FWSVM makes our method model-free as well as computationally efficient due to the piecewise linearity of the FWSVM solutions as functions of the weight. Numerical investigation to both synthetic and real data show the advantageous performance of the proposed method.