Stephen Afrifa^1,2,*, Vijayakumar Varadarajan^3,4,5,*, Peter Appiahene², Tao Zhang¹, Richmond Afrifa⁶

Computer Systems Science and Engineering, Vol.48, No.6, pp. 1625-1639, 2024, DOI:10.32604/csse.2024.055084 - 22 November 2024

Abstract Users of social networks can readily express their thoughts on websites like Twitter (now X), Facebook, and Instagram. The volume of textual data flowing from users has greatly increased with the advent of social media in comparison to traditional media. For instance, using natural language processing (NLP) methods, social media can be leveraged to obtain crucial information on the present situation during disasters. In this work, tweets on the Uttarakhand flash flood are analyzed using a hybrid NLP model. This investigation employed sentiment analysis (SA) to determine the people’s expressed negative attitudes regarding the disaster. More >

Farida Asriani^1,2, Azhari Azhari^1,*, Wahyono Wahyono¹

CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 3079-3096, 2024, DOI:10.32604/cmc.2024.058193 - 18 November 2024

Abstract Incredible progress has been made in human action recognition (HAR), significantly impacting computer vision applications in sports analytics. However, identifying dynamic and complex movements in sports like badminton remains challenging due to the need for precise recognition accuracy and better management of complex motion patterns. Deep learning techniques like convolutional neural networks (CNNs), long short-term memory (LSTM), and graph convolutional networks (GCNs) improve recognition in large datasets, while the traditional machine learning methods like SVM (support vector machines), RF (random forest), and LR (logistic regression), combined with handcrafted features and ensemble approaches, perform well but… More >

Trong Thua Huynh^*, Hoang Thanh Nguyen, Du Thang Phu

CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 2281-2298, 2024, DOI:10.32604/cmc.2024.057954 - 18 November 2024

Abstract In recent years, early detection and warning of fires have posed a significant challenge to environmental protection and human safety. Deep learning models such as Faster R-CNN (Faster Region based Convolutional Neural Network), YOLO (You Only Look Once), and their variants have demonstrated superiority in quickly detecting objects from images and videos, creating new opportunities to enhance automatic and efficient fire detection. The YOLO model, especially newer versions like YOLOv10, stands out for its fast processing capability, making it suitable for low-latency applications. However, when applied to real-world datasets, the accuracy of fire prediction is… More >

Jian Feng^*, Tian Liu, Cailing Du

CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 2895-2909, 2024, DOI:10.32604/cmc.2024.056434 - 18 November 2024

Abstract Unsupervised learning methods such as graph contrastive learning have been used for dynamic graph representation learning to eliminate the dependence of labels. However, existing studies neglect positional information when learning discrete snapshots, resulting in insufficient network topology learning. At the same time, due to the lack of appropriate data augmentation methods, it is difficult to capture the evolving patterns of the network effectively. To address the above problems, a position-aware and subgraph enhanced dynamic graph contrastive learning method is proposed for discrete-time dynamic graphs. Firstly, the global snapshot is built based on the historical snapshots… More >

Gajjala Savithri^1,2, N. Raghavendra Sai^1,*

CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 2547-2569, 2024, DOI:10.32604/cmc.2024.055827 - 18 November 2024

Abstract In the world of wireless sensor networks (WSNs), optimizing performance and extending network lifetime are critical goals. In this paper, we propose a new model called DTLR-Net (Deep Temporal LSTM Regression Network) that employs long-short-term memory and is effective for long-term dependencies. Mobile sinks can move in arbitrary patterns, so the model employs long short-term memory (LSTM) networks to handle such movements. The parameters were initialized iteratively, and each node updated its position, mobility level, and other important metrics at each turn, with key measurements including active or inactive node ratio, energy consumption per cycle,… More >

Shaoxiong Wu¹, Ruoxin Li¹, Xiaofeng Tao¹, Hailong Wu^1,*, Ping Miao¹, Yang Lu¹, Yanyan Lu¹, Qi Liu², Li Pan²

CMC-Computers, Materials & Continua, Vol.81, No.2, pp. 3063-3077, 2024, DOI:10.32604/cmc.2024.055381 - 18 November 2024

Abstract Time series prediction has always been an important problem in the field of machine learning. Among them, power load forecasting plays a crucial role in identifying the behavior of photovoltaic power plants and regulating their control strategies. Traditional power load forecasting often has poor feature extraction performance for long time series. In this paper, a new deep learning framework Residual Stacked Temporal Long Short-Term Memory (RST-LSTM) is proposed, which combines wavelet decomposition and time convolutional memory network to solve the problem of feature extraction for long sequences. The network framework of RST-LSTM consists of two More >

Hui Huang^1,*, Yongbing Li¹, Shuhui Li¹, Ninshu Ma²

The International Conference on Computational & Experimental Engineering and Sciences, Vol.30, No.3, pp. 1-1, 2024, DOI:10.32604/icces.2024.011348

Abstract The sequentially coupled thermo-mechanical finite element analysis (FEA) with implicit iteration scheme is widely adopted for welding process simulation because the one-way coupling scheme is believed to be more efficient. However, such computational framework faces the bottleneck of scalability in large-scale analysis due to the exponential growth of computational burden with respect to the number of unknowns in a FEA model. In the present study, a fully coupled approach with explicit integration was developed to simulate fusion welding induced temperature, distortion, and residual stresses. A mass scaling and heat capacity inverse scaling technique was proposed More >

Ziyu Xu¹, Kuo Tian^1,*

The International Conference on Computational & Experimental Engineering and Sciences, Vol.30, No.2, pp. 1-1, 2024, DOI:10.32604/icces.2024.011245

Abstract Considering experimental testing data is costly, and sensor data is often sparse, while simulation analysis provides overall strength information with lower accuracy, a digital twin framework is proposed for full-field structural strength assessment and prediction. The framework is mainly divided into two stages. In the offline stage, the simulation model of the structure is established, and the sensor layouts are completed. Then, the DNN pre-training model is constructed based on the reduced simulation data. In the online stage, the experimentally measured data are predicted to obtain the time-series sensors data, and the traditional transfer learning… More >

Binbin Hao¹, Zhenming Liu^1,*, Yajun Deng^1,*, Dongliang Sun¹, Wei Zhang¹, Bo Yu¹

The International Conference on Computational & Experimental Engineering and Sciences, Vol.30, No.1, pp. 1-1, 2024, DOI:10.32604/icces.2024.012194

Abstract The tubular fixed bed reactor is widely used in industrial production because of its strong applicability, high stability and easy maintenance. The flow dead zone in the shell of the reactor will significantly affect the overall performance of the reactor. Reducing the flow dead zone in the shell is the main way to optimize the performance of tubular fixed bed reactor. At present, most of the research on the flow dead zone of the reactor is based on the simplified reactor model, the number and size of tubes are far from the industrial requirements. In… More >

Yaren Aydın¹, Gebrail Bekdaş^1,*, Sinan Melih Nigdeli¹, Zong Woo Geem^2,*

CMES-Computer Modeling in Engineering & Sciences, Vol.141, No.3, pp. 2471-2499, 2024, DOI:10.32604/cmes.2024.056693 - 31 October 2024

Abstract Dynamic impacts such as wind and earthquakes cause loss of life and economic damage. To ensure safety against these effects, various measures have been taken from past to present and solutions have been developed using different technologies. Tall buildings are more susceptible to vibrations such as wind and earthquakes. Therefore, vibration control has become an important issue in civil engineering. This study optimizes tuned mass damper inerter (TMDI) using far-fault ground motion records. This study derives the optimum parameters of TMDI using the Adaptive Harmony Search algorithm. Structure displacement and total acceleration against earthquake load More >