• Journal of Information Processing Systems
• /
• v.15 no.5
• /
• pp.1171-1178
• /
• 2019

This study proposes a deep neural network model based on an encoder-decoder structure for visual dialogs. Ongoing linguistic understanding of the dialog history and context is important to generate correct answers to questions in visual dialogs followed by questions and answers regarding images. Nevertheless, in many cases, a visual understanding that can identify scenes or object attributes contained in images is beneficial. Hence, in the proposed model, by employing a separate person detector and an attribute recognizer in addition to visual features extracted from the entire input image at the encoding stage using a convolutional neural network, we emphasize attributes, such as gender, age, and dress concept of the people in the corresponding image and use them to generate answers. The results of the experiments conducted using VisDial v0.9, a large benchmark dataset, confirmed that the proposed model performed well.

• Coupled systems mechanics
• /
• v.10 no.1
• /
• pp.21-38
• /
• 2021

This work is an attempt to design a dynamic model for a non local bio-thermoelastic medium with diffusion. The system of governing equations are formulated in terms of displacement vector field, chemical potential and the tissue temperature in the context of non local dual phase lag (NL DPL) theories of heat conduction and mass diffusion. Based on this considered model, we study the fundamental solution and propagation of plane harmonic waves in tissues. In order to analyze the behavior of the NL DPL model, we construct basic theorem in the terms of elementary function which determine the existence of three longitudinal and one transverse wave. The effects of various parameters on the characteristics of waves i.e., phase velocity and attenuation coefficients are elaborated by plotting various figures of physical quantities in the later part of the paper.

• Journal of information and communication convergence engineering
• /
• v.20 no.1
• /
• pp.31-40
• /
• 2022

Coronary heart disease (CHD) is a comorbidity of COVID-19; therefore, routine early diagnosis is crucial. A large number of examination attributes in the context of diagnosing CHD is a distinct obstacle during the pandemic when the number of health service users is significant. The development of a precise machine learning model for diagnosis with a minimum number of examination attributes can allow examinations and healthcare actions to be undertaken quickly. This study proposes a CHD diagnosis model based on feature selection, data balancing, and ensemble-based classification methods. In the feature selection stage, a hybrid SVM-GA combined with fast correlation-based filter (FCBF) is used. The proposed system achieved an accuracy of 94.60% and area under the curve (AUC) of 97.5% when tested on the z-Alizadeh Sani dataset and used only 8 of 54 inspection attributes. In terms of performance, the proposed model can be placed in the very good category.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.9-16
• /
• 2023

The natural language on social network platforms has a certain front-to-back dependency in structure, and the direct conversion of Chinese text into a vector makes the dimensionality very high, thereby resulting in the low accuracy of existing text classification methods. To this end, this study establishes a deep learning model that combines a big data ultra-deep convolutional neural network (UDCNN) and long short-term memory network (LSTM). The deep structure of UDCNN is used to extract the features of text vector classification. The LSTM stores historical information to extract the context dependency of long texts, and word embedding is introduced to convert the text into low-dimensional vectors. Experiments are conducted on the social network platforms Sogou corpus and the University HowNet Chinese corpus. The research results show that compared with CNN + rand, LSTM, and other models, the neural network deep learning hybrid model can effectively improve the accuracy of text classification.

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.685-691
• /
• 2009

Recently, public office building projects are being recognized by many construction engineers and researchers, as the critical projects in the construction industry. The project budgets have sometimes exceeded due to the lack of core knowledge, experiences, skills and experts concerned in cost planning and estimating in the pre-construction stage. It has been highlighted that planning and estimating effectively the cost of public office building projects as critical in the design stage. Within this context, some cost data books and systems, such as RSMeans cost data systems and Spon's price book, have been systematically developed and used by many construction cost managers and organizations in order to effectively estimate and use their project budgets. As a result of this research, a cost estimating model for finishes has been developed, considering the cost data used in public office building projects.

• International conference on construction engineering and project management
• /
• 2015.10a
• /
• pp.588-592
• /
• 2015

Many studies have affirmed a negative influence of fragmentation on learning and knowledge sharing in construction projects. However, the literature overlooked enablers of learning within this context. The purpose of this paper is to explore the factors that facilitate project learning and ways to negate any unbecoming effects of fragmentation. Qualitative study used to explore the enablers through interviews administered to 11 top management individuals working in different construction projects in Malaysia. The findings revealed the following factors: participation, relationships, togetherness, and roles of project leader and coordinator. The role of boundary objects was also highlighted including information technology (IT), contract and procedures, drawings, specifications, and reports. The outcome of this paper initiates the development of a model for better knowledge creation and sharing in construction projects. The significance of this model stems from its ability to connection both the characteristics of construction project and project learning theories using the enablers. It is envisaged that future work will be to confirm the model in a quantitative study.

• Asia pacific journal of information systems
• /
• v.32 no.2
• /
• pp.354-380
• /
• 2022

This study proposed a novel model to investigate influential factors affecting the intention to continue using increasingly popular food delivery apps (FDAs). The proposed theoretical model is developed and validated to extend traditional technology acceptance and adoption theories by identifying several determinant factors that capture the unique context of FDAs continuous usage. Hypotheses were tested using a partial least square structural equation modeling approach (PLS-SEM) on data collected from 331 actual FDAs users during the COVID-19 pandemic. The results reveal that convenience, perceived compatibility, delivery experience, and online reviews significantly influence the continuous usage of FDAs. The findings also confirm the importance of continuous intention on the actual use of FDAs. The research model of this study explains 65% of variance in continuous intention and 47% in actual use. The insights provided by this study suggest fruitful directions for future research. They can also help FDAs companies, developers and marketers with strategies and tips for further development and growth by ensuring users' continuous usage of these platforms.

• Geomechanics and Engineering
• /
• v.36 no.1
• /
• pp.19-26
• /
• 2024

Erigen's nonlocal thermoelasticity model is used to study the effect of viscosity on a micropolar thermoelastic solid in the context of the multi-phase-lag model. The harmonic wave analysis technique is employed to convert partial differential equations to ordinary differential equations to get the solution to the problem. The physical fields have been presented graphically for the nonlocal micropolar thermoelastic solid. Comparisons are made with the results of three theories different in the presence and absence of viscosity as well as the gravity field. Comparisons are made with the results of three theories different for different values of the nonlocal parameter. Numerical computations are carried out with the help of Matlab software.

• Structural Engineering and Mechanics
• /
• v.89 no.4
• /
• pp.375-381
• /
• 2024

In this paper, Eringen's nonlocal thermoelasticity is constructed to study wave propagation in a rotating two-temperature thermoelastic half-space. The problem is applied in the context of the dual-phase-lag (Dual) model, coupled theory (CD), and Lord-Shulman (L-S) theory. Using suitable non-dimensional fields, the harmonic wave analysis is used to solve the problem. Comparisons are carried with the numerical values predicted in the absence and presence of the gravity field, a nonlocal parameter as well as rotation. The present study is valuable for the analysis of nonlocal thermoelastic problems under the influence of the gravity field, mechanical force, and rotation.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.80-87
• /
• 2024

This paper focuses on improving accuracy in constrained computing settings by employing the ReLU (Rectified Linear Unit) activation function. The research conducted involves modifying parameters of the ReLU function and comparing performance in terms of accuracy and computational time. This paper specifically focuses on optimizing ReLU in the context of a Multilayer Perceptron (MLP) by determining the ideal values for features such as the dimensions of the linear layers and the learning rate (Ir). In order to optimize performance, the paper experiments with adjusting parameters like the size dimensions of linear layers and Ir values to induce the best performance outcomes. The experimental results show that using ReLU alone yielded the highest accuracy of 96.7% when the dimension sizes were 30 - 10 and the Ir value was 1. When combining ReLU with the Adam optimizer, the optimal model configuration had dimension sizes of 60 - 40 - 10, and an Ir value of 0.001, which resulted in the highest accuracy of 97.07%.