Article
Computer Science, Artificial Intelligence
D. B. Jagannadha Rao, Vijayakumar Polepally, S. Nagendra Prabhu, Parsi Kalpana
Summary: This paper proposes a particle squirrel search optimisation-based deep recurrent neural network (PSSO-based DRNN) to predict the coronavirus epidemic (COVID). The cloud-based Hadoop framework is used for the prediction process, involving the mapper and reducer phases. Technical indicators are extracted from the time series data, and feature selection is done using the deep belief network (DBN). The COVID prediction is then made by the DRNN classifier trained using the PSSO algorithm. The proposed method achieves minimal MSE and RMSE for affected, death, and recovered cases.
INTERNATIONAL JOURNAL OF BIO-INSPIRED COMPUTATION
(2023)
Article
Engineering, Electrical & Electronic
Shin Kamada, Takumi Ichimura
Summary: An adaptive structural learning method for restricted Boltzmann machine (RBM) and deep belief network (DBN) has been developed in this study. The algorithm optimizes the network structure by the neuron generation-annihilation algorithm in RBM and the layer generation algorithm in DBN. The proposed method is applied to an automatic road network recognition system called RoadTracer and achieves significantly improved detection accuracy. Furthermore, the method is also applied to the detection of available roads after a natural disaster for rapid transportation retrieval.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Computer Science, Artificial Intelligence
Omar Haddad, Fethi Fkih, Mohamed Nazih Omri
Summary: Cloud computing plays a crucial role in storing and processing Big Data, and deep learning has been widely adopted to extract predictions from user opinions. This paper introduces a new prediction approach, PABIDDL, based on Big Data analysis and deep learning, which achieves high performance on large-scale datasets.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
S. T. Suganthi, Mohamed Uvaze Ahamed Ayoobkhan, Krishna Kumar, Nebojsa Bacanin, K. Venkatachalam, Hubalovsky Stepan, Trojovsky Pavel
Summary: This article discusses the application of deep learning in detecting deep fake images, identifies the issues with existing techniques, and proposes a method that combines Fisherface and LBPH algorithms for deep fake detection in face images.
PEERJ COMPUTER SCIENCE
(2022)
Article
Engineering, Electrical & Electronic
Yunpeng Gao, Yunfeng Li, Yanqing Zhu, Cong Wu, Dexi Gu
Summary: A new method combining adaptive wavelet threshold denoising and deep belief network fusion extreme learning machine (DBN-ELM) is proposed to solve the problems of noise interference and artificial feature extraction in power quality disturbance (PQD) classification. The simulation result and experimental verification show that the proposed method can effectively suppress PQD noise and performs well on DBN-ELM classification.
ELECTRIC POWER SYSTEMS RESEARCH
(2022)
Article
Engineering, Electrical & Electronic
Xiaofeng Yuan, Yongjie Gu, Yalin Wang
Summary: A novel supervised DBN (SDBN) is proposed in this article by introducing quality information into the training phase, ensuring learned features are largely quality-related for soft sensor modeling. The SDBN-based soft sensor model shows improved performance in quality prediction for industrial processes such as a debutanizer column and a hydrocracking process.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Computer Science, Theory & Methods
Tanvi Chawla, Girdhari Singh, Emmanuel S. Pilli
Summary: The paper introduces an efficient distributed RDF storage scheme called MuSe for storing and querying RDF data with Hadoop MapReduce. MuSe optimizes RDF storage to answer frequently occurring triple patterns in minimum time and outperforms compared frameworks in terms of query execution time and scalability, as demonstrated by experiments on synthetic RDF datasets LUBM and WatDiv.
JOURNAL OF BIG DATA
(2021)
Article
Computer Science, Artificial Intelligence
Burcu Kir Savas, Yasar Becerikli
Summary: Traffic accidents caused by driver fatigue and drowsiness have resulted in numerous injuries and deaths. Therefore, driver fatigue detection and prediction systems have been recognized as important research areas to prevent such accidents. This study proposes the use of a deep belief network (DBN) model for classifying fatigue symptoms, achieving a high accuracy rate of approximately 86% in experimental tests.
NEURAL COMPUTING & APPLICATIONS
(2022)
Review
Physics, Multidisciplinary
Aurelien Decelle, Cyril Furtlehner
Summary: This review focuses on the application of restricted Boltzmann machines in statistical physics, discussing mean-field theory and phase diagram analysis of RBMs in machine learning. Recent works on mean-field based learning algorithms and reproducing aspects of the learning process from ensemble dynamics equations or linear stability arguments are also discussed.
Article
Computer Science, Artificial Intelligence
Yuhua Chen, Hasri Mustafa, Xuandong Zhang, Jing Liu
Summary: In order to improve the information application level of financial management and the business value of financial big data, this article automatically classifies financial data using the fuzzy clustering algorithm and detects abnormal data using the local outlier factor (LOF) algorithm based on neighborhood relation. A financial data management platform based on distributed Hadoop architecture is designed, combining MapReduce framework with the fuzzy clustering algorithm and LOF algorithm to enhance the algorithm's performance and accuracy, thus improving operational efficiency of enterprise financial data processing. Comparative experimental results demonstrate that the proposed platform achieves the best running efficiency and financial data classification accuracy compared with other methods, illustrating the effectiveness and superiority of the platform.
PEERJ COMPUTER SCIENCE
(2023)
Article
Computer Science, Artificial Intelligence
Carolina Salto, Gabriela Minetti, Enrique Alba, Gabriel Luque
Summary: This article discusses the use of MapReduce as a computing paradigm to solve large-scale combinatorial optimization problems, focusing on the potential and advantages of developing genetic algorithms using Hadoop, Spark, and MPI as middleware platforms. The results show that MRGA performs better on the Hadoop framework compared to Spark and MPI when dealing with high-dimensional datasets.
Article
Construction & Building Technology
Ahmed Meshref, Karim El-Dash, Mohamed Basiouny, Omia El-Hadidi
Summary: This article presents an LCC deep learning prediction model to assess structural and envelope-type alternatives for industrial building and make the most suitable structure decision. The model is validated through investigation cases and comparison of design structure alternatives, concluding that precast/pre-stressed concrete frames are the best choice.
Article
Construction & Building Technology
Feng Shangxin, Chen Zuyu, Luo Hua, Wang Shanyong, Zhao Yufei, Liu Lipeng, Ling Daosheng, Jing Liujie
Summary: This study explores the potential of using deep learning to predict TBM performance in a 17.5 km tunnel excavated for the Yingsong Water Diversion Project in Northeastern China. By introducing Field Penetration Index (FPI) to quantify TBM performance, the deep belief network (DBN) is utilized to predict performance with good agreement with field data. Results confirm the usefulness of big data and deep learning in predicting TBM performance.
TUNNELLING AND UNDERGROUND SPACE TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Hamidreza Kadkhodaei, Amir Masoud Eftekhari Moghadam, Mehdi Dehghan
Summary: This paper presents a distributed heterogeneous ensemble classifier for big data, which utilizes multiple classifiers to achieve more accurate data classification. Experimental results indicate the superiority of the proposed method in terms of classification accuracy, performance, and scalability compared to existing ensemble algorithms.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Dai-Lun Chiang, Sheng-Kuan Wang, Yu-Ying Wang, Yi-Nan Lin, Tsang-Yen Hsieh, Cheng-Ying Yang, Victor R. L. Shen, Hung-Wei Ho
Summary: This paper explores using a Petri net to create a visual model of the MapReduce framework and analyze its reachability property, demonstrating the feasibility of a real big data analysis system and proposing an error prevention mechanism to increase development efficiency.
APPLIED ARTIFICIAL INTELLIGENCE
(2021)