10 results on '"Liu, Weibo"'
Search Results
2. A Novel Ensemble-Learning-Based Convolution Neural Network for Handling Imbalanced Data
- Author
-
Wu, Xianbin, Wen, Chuanbo, Wang, Zidong, Liu, Weibo, and Yang, Junjie
- Published
- 2024
- Full Text
- View/download PDF
3. Novel particle swarm optimization algorithms with applications to healthcare data analysis
- Author
-
Liu, Weibo, Wang, Z., and Liu, X.
- Subjects
006.3 ,Evolutionary computation ,Deep learning ,Data analysis ,Clustering ,Classification - Abstract
Optimization problem is a fundamental research topic which has been receiving increasing interest according to its application potential in almost all real-world systems including engineering systems, large-scaled complex networks, healthcare management systems and so on. A large number of heuristic algorithms have been developed with the purpose of effectively solving the optimization problems during the past few decades. Served as a powerful family of heuristic algorithms, the particle swarm optimization (PSO) algorithm has been successfully employed in a variety of practical applications in dealing with optimization problems. The PSO algorithm has exhibited more competitive performance than many popular evolutionary computation approaches because of its easy implementation, fast convergence and comprehensive ability of converging to a satisfactory solution. Nevertheless, there is still much room to improve the PSO algorithm in terms of both the convergence rate and the population diversity. To summarize, there are three challenging problems in developing new variant PSO algorithms with hope to further improve the convergence rate of the PSO algorithm and maintain the population diversity: 1) how to adjust the control parameters of the PSO algorithm; 2) how to achieve the balance between the local search and the global search during the evolution process; and 3) how to guarantee the search ability of the particles and avoid premature convergence. In this thesis, we address the above mentioned challenging problems and aim to design effective variant PSO algorithms with applications in intelligent data analysis. It should be pointed out that all the developed PSO algorithms in this thesis have been evaluated by comparing with some currently popular variant PSO algorithms. • With the aim to improve the convergence rate of the optimizer, an adaptive weighting PSO algorithm is put forward where a sigmoid-function-based weighting strategy is introduced to adjust the acceleration coefficients. With this weighting strategy, the distances from the particle to the global best position and from the particle to its personal best position are both taken into consideration, thereby having the distinguishing feature of enhancing the convergence rate. • As with other evolutionary computation approaches, the modification of parameters is an efficient method for improving the search ability of the algorithm. We present a randomised PSO algorithm where Gaussian white noise with adjustable intensity is utilized to randomly perturb the acceleration coefficients in order to explore and exploit the problem space thoroughly. • To further develop a novel PSO algorithm with promising search ability, we propose a randomly occurring distributedly delayed particle swarm optimization (RODDPSO) algorithm which demonstrates competitive performance in seeking the optimal solution. The randomly occurring distributed time delays not only contribute to a thorough exploration of the search space but also achieve a proper balance between the local exploitation and the global exploration. • To fully investigate the application potential of the developed PSO algorithms, we apply the RODDPSO algorithm to intelligent data analysis (including data clustering and classification problems). We optimize the initial cluster centroids of the K-means clustering algorithm and the hyperparameters of the deep neural network by using the RODDPSO algorithm. The developed PRODDPSO-based algorithms are successfully employed in patients’ triage categorization and patient attendance disposal problems with satisfactory performance.
- Published
- 2020
4. A PSO-based deep learning approach to classifying patients from emergency departments
- Author
-
Liu, Weibo, Wang, Zidong, Zeng, Nianyin, Alsaadi, Fuad E., and Liu, Xiaohui
- Published
- 2021
- Full Text
- View/download PDF
5. A survey of deep neural network architectures and their applications.
- Author
-
Liu, Weibo, Wang, Zidong, Liu, Xiaohui, Zeng, Nianyin, Liu, Yurong, and Alsaadi, Fuad E.
- Subjects
- *
ARTIFICIAL neural networks , *DEEP learning , *PATTERN recognition systems , *AUTOMATIC speech recognition , *NATURAL language processing - Abstract
Since the proposal of a fast learning algorithm for deep belief networks in 2006, the deep learning techniques have drawn ever-increasing research interests because of their inherent capability of overcoming the drawback of traditional algorithms dependent on hand-designed features. Deep learning approaches have also been found to be suitable for big data analysis with successful applications to computer vision, pattern recognition, speech recognition, natural language processing, and recommendation systems. In this paper, we discuss some widely-used deep learning architectures and their practical applications. An up-to-date overview is provided on four deep learning architectures, namely, autoencoder, convolutional neural network, deep belief network, and restricted Boltzmann machine. Different types of deep neural networks are surveyed and recent progresses are summarized. Applications of deep learning techniques on some selected areas (speech recognition, pattern recognition and computer vision) are highlighted. A list of future research topics are finally given with clear justifications. [ABSTRACT FROM AUTHOR]
- Published
- 2017
- Full Text
- View/download PDF
6. Collaborative filtering via heterogeneous neural networks.
- Author
-
Zeng, Wei, Fan, Ge, Sun, Shan, Geng, Biao, Wang, Weiyi, Li, Jiacheng, and Liu, Weibo
- Subjects
ARTIFICIAL neural networks ,COMPUTER vision ,MATRIX decomposition - Abstract
Over the last few years, the deep neural network is utilized to solve the collaborative filtering problem, a method of which has achieved immense success on computer vision, speech recognition as well as natural language processing. On one hand, the deep neural network can be used to capture the side information of users and items. On the other hand, it is also capable of modeling interactions between users and items. Most of existing approaches exploit the neural network with solo structure to model user–item interactions such that the learning representation may be insufficient over the extremely sparse rating data. Recently, a large number of neural networks with mixed structures are devised for learning better representations. A carefully designed hybrid network is able to achieve considerable accuracy but only requires a small amount of extra computation. In order to model user–item interactions, we elaborate a hybrid neural network consisting of the global neural network and several local neural blocks. The multi-layer perceptron is adopted to build the global neural network and the residual network is used to form the local neural block which is inserted into two adjacent global layers. The hybrid network is further combined with the generalized matrix factorization to capture both the linear and nonlinear relationships between users and items. It is verified by experimental results on benchmark datasets that our method is superior to certain state-of-the-art approaches in terms of top-n item recommendation. • Devise a hybrid neural network consisting of a global network and local blocks. • Apply the hybrid neural network to model user-item interactions. • Verify the hybrid network superiority in solving CF by comparing with rivals. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
7. A new particle-swarm-optimization-assisted deep transfer learning framework with applications to outlier detection in additive manufacturing.
- Author
-
Fang, Jingzhong, Wang, Zidong, Liu, Weibo, Chen, Linwei, and Liu, Xiaohui
- Subjects
- *
OUTLIER detection , *DEEP learning , *MARGINAL distributions , *WELDING equipment , *ELECTRIC arc - Abstract
In wire arc additive manufacturing (WAAM), the electric arc is an essential part of the welding equipment, which serves as the heat source and is directed by the current and voltage. The working status of the electric arc is an important factor in determining the quality of the fabricated components. During the welding process, the current and voltage may change abruptly due to some abnormalities in the operating conditions, which may affect the working status and thereby affect the quality of products. Such abnormal changes in the current and voltage can be treated as outliers. In order to identify outliers in current and voltage to further improve the welding process, in this paper, a novel deep-transfer-learning-embedded outlier detection approach is developed for WAAM. A new domain adaptation strategy is designed where the cross-domain discrepancies of the marginal distribution and conditional distribution are minimized. Specifically, two separate coefficients are introduced to adjust the conditional domain discrepancies of normal instances and outliers with the purpose of alleviating the data imbalance problem. The particle swarm optimizer is employed to adjust the hyper-parameters. The developed deep transfer learning framework is exploited in designing a new outlier detector with application to WAAM. The proposed approach is exploited in real-world industrial data collected through the WAAM process. Experimental results demonstrate that the proposed outlier detection approach outperforms the standard deep-learning-based outlier detector approach and the standard transfer-learning-embedded outlier detection approach in terms of detection accuracy. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
8. Bearing fault diagnosis via fusing small samples and training multi-state Siamese neural networks.
- Author
-
Wen, Chuanbo, Xue, Yipeng, Liu, Weibo, Chen, Guochu, and Liu, Xiaohui
- Subjects
- *
ARTIFICIAL neural networks , *FAULT diagnosis , *DEEP learning , *FEATURE extraction , *LIFTING & carrying (Human mechanics) - Abstract
Recently, deep learning techniques have been widely applied to fault diagnosis due to their outstanding feature extraction abilities. The success of deep-learning-based fault diagnosis methods is highly dependent on the quantity and quality of the training data. In practical scenarios, it is challenging to obtain sufficient high-quality training data for fault diagnosis tasks due to complex environments, which would affect the effectiveness of the deep learning methods. In this paper, a new fault diagnosis method is proposed for motor bearing fault diagnosis under small samples. The Siamese neural networks (SNNs) are employed to extract the fault features. A multi-stage training strategy is proposed to train the SNNs with the aim to prevent the training stagnation problem and handle the small sample problem. A multi-source feature fusion network is developed to make full use of the multi-source sensor data by fusing the extracted fault features for further fault diagnosis. The proposed method is applied to motor bearing fault diagnosis on two real-world datasets. Experimental results demonstrate the effectiveness and feasibility of the introduced method for motor bearing fault diagnosis under small samples. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
9. A novel framework for motor bearing fault diagnosis based on multi-transformation domain and multi-source data.
- Author
-
Xue, Yipeng, Wen, Chuanbo, Wang, Zidong, Liu, Weibo, and Chen, Guochu
- Abstract
Through the application of deep learning and multi-sensor data, fault features can be automatically extracted and valuable information can be integrated to tackle intricate challenges in motor bearing fault diagnosis. Most existing fusion models focus primarily on the original time series signal with information extraction largely restricted to the time domain (without extensions into multiple transformation domains). Also, in most fusion models, the sensor fusion level is kept relatively simple which could lead to the oversight of correlations and complementarities among the information. To enhance the recognition capability of diagnostic network features, in this paper, we propose a novel framework for motor bearing fault diagnosis from the perspectives of multi-transformation domain and multi-source data fusion. Within this framework, feature extraction and fusion from various source data are achieved in the time domain, frequency domain, and time–frequency domain. Distinct independent networks are set up within these domains: one network is designated for overseeing feature fusion, while the others are dedicated to extracting features from individual sensors. To support the extraction of pivotal features across multiple fusion layers in various transformation domains, several fusion nodes are inserted between the layers of the multiple feature extraction networks and the feature summarization network. Furthermore, a channel attention mechanism is introduced as a fusion strategy that serves to pinpoint the significance of different features, thus enhancing the efficiency of feature extraction. Experimental evaluation reveals the efficacy of the proposed model and highlights its noteworthy performance attributes such as scalability and universality. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF
10. Deep-reinforcement-learning-based images segmentation for quantitative analysis of gold immunochromatographic strip.
- Author
-
Zeng, Nianyin, Li, Han, Wang, Zidong, Liu, Weibo, Liu, Songming, Alsaadi, Fuad E., and Liu, Xiaohui
- Subjects
- *
QUANTITATIVE research , *IMAGE analysis , *REINFORCEMENT learning , *IMMUNOASSAY , *DEEP learning , *GOLD , *IMAGE segmentation - Abstract
Gold immunochromatographic strip (GICS) is a widely used lateral flow immunoassay technique. A novel image segmentation method is developed in this paper for quantitative analysis of GICS based on the deep reinforcement learning (DRL), which can accurately distinguish the test line and the control line in the GICS images. The deep belief network (DBN) is employed in the deep Q network in our DRL algorithm. Meanwhile, the multi-factor learning curve is introduced in the DRL algorithm to dynamically adjust the capacity of the replay buffer and the sampling size, which leads to enhanced learning efficiency. It is worth mentioning that the states, actions, and rewards in the developed DRL algorithm are determined based on the characteristics of GICS images. Experiment results demonstrate the feasibility and reliability of the proposed DRL-based image segmentation method and show that the proposed new image segmentation method outperforms some existing image segmentation methods for quantitative analysis of GICS images. [ABSTRACT FROM AUTHOR]
- Published
- 2021
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.