• Advances in nano research
• /
• 제15권6호
• /
• pp.533-539
• /
• 2023

This study aims to examine four machine learning (ML)-based models for their potential to estimate the splitting tensile strength (STS) of manufactured sand concrete (MSC). The ML models were trained and tested based on 310 experimental data points. Stone nanopowder content (SNPC), curing age (CA), and water-to-cement (W/C) ratio were also studied for their impacts on the STS of MSC. According to the results, the support vector regression (SVR) model had the highest correlation with experimental data. Still, all of the optimized ML models showed promise in estimating the STS of MSC. Both ML and laboratory results showed that MSC with 10% SNPC improved the STS of MSC.

• Computers and Concrete
• /
• 제17권3호
• /
• pp.353-386
• /
• 2016

Realistic assessment of the performance of reinforced concrete structural members like columns is needed for designing new structures or maintenance of the existing structural members. This assessment requires analytical capability of employing proper material models and cyclic rules and considering various load and displacement patterns. A computer application was developed to analyze the non-linear, cyclic flexural performance of reinforced concrete structural members under various types of loading paths including non-sequential variations in axial load and bi-axial cyclic load or displacement. Different monotonic material models as well as hysteresis rules, were implemented in a fiber-based moment-curvature and in turn force-deflection analysis, using proper assumptions on curvature distribution along the member, as in plastic-hinge models. Performance of the program was verified against analytical results by others, and accuracy of the analytical process and the implemented models were evaluated in comparison to the experimental results. The computer application can be used to predict the response of a member with an arbitrary cross section and various type of lateral and longitudinal reinforcement under different combinations of loading patterns in axial and bi-axial directions. On the other hand, the application can be used to examine analytical models and methods using proper experimental data.

• Journal of Computing Science and Engineering
• /
• 제9권1호
• /
• pp.9-19
• /
• 2015

The most important challenge in the region-based abstraction method as an approach to compute the state space of time Petri Nets (TPNs) for model checking is that the method results in a huge number of regions, causing a state explosion problem. Thus, region-based abstraction methods are not appropriate for use in developing practical tools. To address this limitation, this paper applies a modification to the basic region abstraction method to be used specially for computing the state space of TPN models, so that the number of regions becomes smaller than that of the situations in which the current methods are applied. The proposed approach is based on the special features of TPN that helps us to construct suitable and small region graphs that preserve the time properties of TPN. To achieve this, we use TPN-TCTL as a timed extension of CTL for specifying a subset of properties in TPN models. Then, for model checking TPN-TCTL properties on TPN models, CTL model checking is used on TPN models by translating TPN-TCTL to the equivalent CTL. Finally, we compare our proposed method with the current region-based abstraction methods proposed for TPN models in terms of the size of the resulting region graph.

• Journal of Communications and Networks
• /
• 제15권2호
• /
• pp.228-239
• /
• 2013

Peer-to-peer (P2P) systems have proved the most effective and popular file sharing applications in recent years. Previous studies mainly focused on equal service and differentiated service strategies when peers have no initial data before their downloads. For an upload-constrained P2P file sharing system, we model both the equal service process and the differentiated service process when the initial data distribution of peers satisfies some special conditions. Moreover, we show how to minimize the time required to distribute the file to any number of peers. The proposed fluid-based models can reveal the intrinsic relations among the initial data amount, the peer set size, and the minimum last finish time. The closed-form expressions derived from the extended models can closely approximate chunk-based models and systems, especially for relatively large files. As an application of the extended models, we show how to provide differentiated service efficiently to multiple peer sets. Since no limits are imposed on the upload bandwidth of peers or the size of each peer set, we believe that our analytic process and the results achieved can provide not only fundamental insights into bandwidth allocation and data scheduling but also a helpful reference for both improving system performance and building an effective incentive mechanism for P2P file sharing systems.

• International Journal of Reliability and Applications
• /
• 제17권2호
• /
• pp.149-158
• /
• 2016

We are enjoying a very comfortable life thanks to modern civilization, however, comfort is not guaranteed to us. Development of software system is a difficult and complex process. Therefore, the main focus of software development is on improving the reliability and stability of a software system. We have become aware of the importance of developing software reliability models and have begun to develop software reliability models. NHPP software reliability models have been developed through the fault intensity rate function and the mean value functions within a controlled testing environment to estimate reliability metrics such as the number of residual faults, failure rate, and reliability of the software. In this paper, we present a new NHPP software reliability model with Burr Type III fault detection rate, and present the goodness-of-fit of the fault detection rate software reliability model and other NHPP models based on two datasets of software testing data. The results show that the proposed model fits significantly better than other NHPP software reliability models.

• Geomechanics and Engineering
• /
• 제37권4호
• /
• pp.307-321
• /
• 2024

This paper delves into the critical assessment of predicting sidewall displacement in underground caverns through the application of nine distinct machine learning techniques. The accurate prediction of sidewall displacement is essential for ensuring the structural safety and stability of underground caverns, which are prone to various geological challenges. The dataset utilized in this study comprises a total of 310 data points, each containing 13 relevant parameters extracted from 10 underground cavern projects located in Iran and other regions. To facilitate a comprehensive evaluation, the dataset is evenly divided into training and testing subset. The study employs a diverse array of machine learning models, including recurrent neural network, back-propagation neural network, K-nearest neighbors, normalized and ordinary radial basis function, support vector machine, weight estimation, feed-forward stepwise regression, and fuzzy inference system. These models are leveraged to develop predictive models that can accurately forecast sidewall displacement in underground caverns. The training phase involves utilizing 80% of the dataset (248 data points) to train the models, while the remaining 20% (62 data points) are used for testing and validation purposes. The findings of the study highlight the back-propagation neural network (BPNN) model as the most effective in providing accurate predictions. The BPNN model demonstrates a remarkably high correlation coefficient (R2 = 0.99) and a low error rate (RMSE = 4.27E-05), indicating its superior performance in predicting sidewall displacement in underground caverns. This research contributes valuable insights into the application of machine learning techniques for enhancing the safety and stability of underground structures.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2005년도 춘계 학술대회 발표논문집
• /
• pp.87-90
• /
• 2005

Today's state-of the-art speech recognition systems typically use continuous distribution hidden Markov models with the mixtures of Gaussian distributions. To obtain higher recognition accuracy, the hidden Markov models typically require huge number of Gaussian distributions. Such speech recognition systems have problems that they require too much memory to run, and are too slow for large applications. Many approaches are proposed for the design of compact acoustic models. One of those models is subspace distribution clustering hidden Markov model. Subspace distribution clustering hidden Markov model can represent original full-space distributions as some combinations of a small number of subspace distribution codebooks. Therefore, how to make the codebook is an important issue in this approach. In this paper, we report some experimental results on various quantization methods to make more accurate models.

• ETRI Journal
• /
• 제44권5호
• /
• pp.794-804
• /
• 2022

Transformer models have had a great impact on natural language processing (NLP) in recent years by realizing outstanding and efficient contextualized language models. Recent studies have used transformer-based language models for various NLP tasks, including Persian named entity recognition (NER). However, in complex tasks, for example, NER, it is difficult to determine which contextualized embedding will produce the best representation for the tasks. Considering the lack of comparative studies to investigate the use of different contextualized pretrained models with sequence modeling classifiers, we conducted a comparative study about using different classifiers and embedding models. In this paper, we use different transformer-based language models tuned with different classifiers, and we evaluate these models on the Persian NER task. We perform a comparative analysis to assess the impact of text representation and text classification methods on Persian NER performance. We train and evaluate the models on three different Persian NER datasets, that is, MoNa, Peyma, and Arman. Experimental results demonstrate that XLM-R with a linear layer and conditional random field (CRF) layer exhibited the best performance. This model achieved phrase-based F-measures of 70.04, 86.37, and 79.25 and word-based F scores of 78, 84.02, and 89.73 on the MoNa, Peyma, and Arman datasets, respectively. These results represent state-of-the-art performance on the Persian NER task.

• Communications for Statistical Applications and Methods
• /
• 제10권1호
• /
• pp.125-134
• /
• 2003

Standard estimation methods for linear mixed models are sensitive to influential observations. However, tools and concepts for linear mixed model diagnostics are rudimentary until now and research is heavily demanded in linear mixed models. In this paper, we consider two diagnostics to evaluate the effects of individual observations in the estimation of fixed effects for linear mixed models. Those are Cook's distance and COVRATIO. Results of our limited simulation study suggest that the Cook's distance is not good statistical quantity in linear mixed models. Also calibration point for COVRATIO seems to be quite conservative.

• Journal of Multimedia Information System
• /
• 제6권4호
• /
• pp.209-216
• /
• 2019

Over the past decade, researchers were able to solve complex medical problems as well as acquire deeper understanding of entire issue due to the availability of machine learning techniques, particularly predictive algorithms and automatic recognition of patterns in medical imaging. In this study, a technique called transfer learning has been utilized to classify Magnetic Resonance (MR) images by a pre-trained Convolutional Neural Network (CNN). Rather than training an entire model from scratch, transfer learning approach uses the CNN model by fine-tuning them, to classify MR images into Alzheimer's disease (AD), mild cognitive impairment (MCI) and normal control (NC). The performance of this method has been evaluated over Alzheimer's Disease Neuroimaging (ADNI) dataset by changing the learning rate of the model. Moreover, in this study, in order to demonstrate the transfer learning approach we utilize different pre-trained deep learning models such as GoogLeNet, VGG-16, AlexNet and ResNet-18, and compare their efficiency to classify AD. The overall classification accuracy resulted by GoogLeNet for training and testing was 99.84% and 98.25% respectively, which was exceptionally more than other models training and testing accuracies.