• Journal of Korean Association for Spatial Structures
• /
• v.22 no.2
• /
• pp.39-46
• /
• 2022

Recently, machine learning is widely used to solve optimization problems in various engineering fields. In this study, machine learning is applied to development of a control algorithm for a smart control device for reduction of seismic responses. For this purpose, Deep Q-network (DQN) out of reinforcement learning algorithms was employed to develop control algorithm. A single degree of freedom (SDOF) structure with a smart tuned mass damper (TMD) was used as an example structure. A smart TMD system was composed of MR (magnetorheological) damper instead of passive damper. Reward design of reinforcement learning mainly affects the control performance of the smart TMD. Various hyper-parameters were investigated to optimize the control performance of DQN-based control algorithm. Usually, decrease of the time step for numerical simulation is desirable to increase the accuracy of simulation results. However, the numerical simulation results presented that decrease of the time step for reward calculation might decrease the control performance of DQN-based control algorithm. Therefore, a proper time step for reward calculation should be selected in a DQN training process.

• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.119-130
• /
• 2022

Crime is a common social problem that affects the quality of life. As the number of crimes increases, it is necessary to build a model to predict the number of crimes that may occur in a given period, identify the characteristics of a person who may commit a particular crime, and identify places where a particular crime may occur. Data privacy is the main challenge that organizations face when building this type of predictive models. Federated learning (FL) is a promising approach that overcomes data security and privacy challenges, as it enables organizations to build a machine learning model based on distributed datasets without sharing raw data or violating data privacy. In this paper, a federated long short- term memory (LSTM) model is proposed and compared with a traditional LSTM model. Proposed model is developed using TensorFlow Federated (TFF) and the Keras API to predict the number of crimes. The proposed model is applied on the Boston crime dataset. The proposed model's parameters are fine tuned to obtain minimum loss and maximum accuracy. The proposed federated LSTM model is compared with the traditional LSTM model and found that the federated LSTM model achieved lower loss, better accuracy, and higher training time than the traditional LSTM model.

• Steel and Composite Structures
• /
• v.44 no.3
• /
• pp.389-406
• /
• 2022

Recycling concrete construction waste is an encouraging step toward green and sustainable building. A lot of research has been done on recycled aggregate concretes (RACs), but not nearly as much has been done on concrete made with recycled aggregate. Recycled aggregate concrete, on the other hand, has been found to have a lower mechanical productivity compared to conventional one. Accurately estimating the mechanical behavior of the concrete samples is a most important scientific topic in civil, structural, and construction engineering. This may prevent the need for excess time and effort and lead to economic considerations because experimental studies are often time-consuming, costly, and troublous. This study presents a comprehensive data-mining-based model for predicting the splitting tensile strength of recycled aggregate concrete modified with glass fiber and silica fume. For this purpose, first, 168 splitting tensile strength tests under different conditions have been performed in the laboratory, then based on the different conditions of each experiment, some variables are considered as input parameters to predict the splitting tensile strength. Then, three hybrid models as GWO-RF, GWO-MLP, and GWO-SVR, were utilized for this purpose. The results showed that all developed GWO-based hybrid predicting models have good agreement with measured experimental results. Significantly, the GWO-RF model has the best accuracy based on the model performance assessment criteria for training and testing data.

• Journal of Information Processing Systems
• /
• v.18 no.4
• /
• 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.

• Advances in aircraft and spacecraft science
• /
• v.9 no.6
• /
• pp.537-551
• /
• 2022

The rapid decrease of fossil fuel resources and increase of environmental pollution caused by aviation industries have become a severe issue which leads to an increase in the greenhouse effect. The use of biofuel becomes an option to alleviate issues related to unrenewable resources. This study presents a computational simulation of the biofuel combustion characteristics of various alternative fuels in an annular combustion chamber designed for training aircraft. The biofuels used in this study are Sorghum Oil Methyl Ester (SOME), Spirulina Platensis Algae (SPA) and Camelina Hydrotreated Esters and Fatty Acids (CHEFA). Meanwhile, Jet-A is used as a baseline fuel. The fuel properties and combustion characteristics are being investigated and analysed. The results are presented in terms of temperature and pressure profiles in addition to the formation of NOx and soot generated from the combustion chamber. Results obtained show that CHEFA fuel is the most recommended biofuel among all four tested fuels as it is being found that it burns with 37.6% lower temperature, 15.2% lower pressure, 89.5% lower NOx emission and 8.1% lower soot emission compared with the baseline fuel in same combustion chamber geometry with same initial parameters.

• Structural Engineering and Mechanics
• /
• v.87 no.6
• /
• pp.517-527
• /
• 2023

The severe deterioration of structures has led to extensive research on the development of structural repair techniques using composite materials. Consequently, previous researchers have devised various analytical methods to predict the interface performance of bonded repairs. However, these analytical solutions are highly complex mathematically and necessitate numerous calculations with a large number of iterations to obtain the output parameters. In this paper, an artificial neural network prediction models is used to calculate the interfacial stress distribution in RC beams strengthened with FRP sheet. The R2value for the training data is evaluated as 0.99, and for the testing data, it is 0.92. Closed-form solutions are derived for RC beams strengthened with composite sheets simply supported at both ends and verified through direct comparisons with existing results. A comparative study of peak interfacial shear and normal stresses with the literature gives the usefulness and effectiveness of ANN proposed. A parametrical study is carried out to show the effects of some design variables, e.g., thickness of adhesive layer and FRP sheet.

• Annual Conference of KIPS
• /
• 2011.04a
• /
• pp.447-450
• /
• 2011

This paper shows the application of P-N Learning [4] method in the soccer ball detection and improvement for increasing the speed of processing. In the P-N learning, the learning process is guided by positive (P) and negative (N) constraints which restrict the labeling of the unlabeled data, identify examples that have been classified in contradiction with structural constraints and augment the training set with the corrected samples in an iterative process. But for the long-view in the soccer game, P-N learning will produce so many ferns that more time is spent than other methods. We propose that color histogram of each frame is constructed to delete the unnecessary details in order to decreasing the number of feature points. We use the mask to eliminate the gallery region and Line Hough Transform to remove the line and adjust the P-N learning's parameters to optimize accurate and speed.

• Communications for Statistical Applications and Methods
• /
• v.29 no.2
• /
• pp.177-191
• /
• 2022

This paper addresses the use of machine learning methods for causal estimation of treatment effects from observational data. Even though conducting randomized experimental trials is a gold standard to reveal potential causal relationships, observational study is another rich source for investigation of exposure effects, for example, in the research of comparative effectiveness and safety of treatments, where the causal effect can be identified if covariates contain all confounding variables. In this context, statistical regression models for the expected outcome and the probability of treatment are often imposed, which can be combined in a clever way to yield more efficient and robust causal estimators. Recently, targeted maximum likelihood estimation and causal random forest is proposed and extensively studied for the use of data-adaptive regression in estimation of causal inference parameters. Machine learning methods are a natural choice in these settings to improve the quality of the final estimate of the treatment effect. We explore how we can adapt the design and training of several machine learning algorithms for causal inference and study their finite-sample performance through simulation experiments under various scenarios. Application to the percutaneous coronary intervention (PCI) data shows that these adaptations can improve simple linear regression-based methods.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.252-255
• /
• 2020

Unsupervised deep learning methods have shown impressive results for the challenging monocular depth estimation task, a field of study that has gained attention in recent years. A common approach for this task is to train a deep convolutional neural network (DCNN) via an image synthesis sub-task, where additional views are utilized during training to minimize a photometric reconstruction error. Previous unsupervised depth estimation networks are trained within a fixed depth estimation range, irrespective of its possible range for a given image, leading to suboptimal estimates. To overcome this suboptimal limitation, we first propose an unsupervised adaptive depth estimation method guided by minimum and maximum (min-max) depth priors for a given input image. The incorporation of min-max depth priors can drastically reduce the depth estimation complexity and produce depth estimates with higher accuracy. Moreover, we propose a novel network architecture for adaptive depth estimation, called the AdaMM-DepthNet, which adopts the min-max depth estimation in its front side. Intensive experimental results demonstrate that the adaptive depth estimation can significantly boost up the accuracy with a fewer number of parameters over the conventional approaches with a fixed minimum and maximum depth range.

• Geomechanics and Engineering
• /
• v.34 no.1
• /
• pp.43-49
• /
• 2023

Knowledge of minimum horizontal stress (S_hmin) is a significant step in determining full stress tensor. It provides crucial information for the production of sand, hydraulic fracturing, determination of safe mud weight window, reservoir production behavior, and wellbore stability. Calculating the S_hmin using indirect methods has been proved to be awkward because a lot of data are required in all of these models. Also, direct techniques such as hydraulic fracturing are costly and time-consuming. To figure these problems out, this work aims to apply the long-short-term memory (LSTM) algorithm to S_hmin time-series prediction. 13956 datasets obtained from an oil well logging operation were applied in the models. 80% of the data were used for training, and 20% of the data were used for testing. In order to achieve the maximum accuracy of the LSTM model, its hyper-parameters were optimized significantly. Through different statistical indices, the LSTM model's performance was compared with with other machine learning methods. Finally, the optimized LSTM model was recommended for S_hmin prediction in the well logging operation.