Entity

Time filter

Source Type


Yan A.,Beijing University of Technology | Yan A.,Beijing Key Laboratory of Computational Intelligence and Intelligent System | Wang D.,La Trobe University
Expert Systems with Applications | Year: 2015

To achieve better classification performance using case-based reasoning classifiers, we propose a retrieval-based revision method with trustworthiness evaluation for problem solving. An improved case evaluation method is employed to evaluate the trustworthiness of the suggested solution after the reuse step, which will divide the target cases and its suggested solutions into a trustworthy set and an untrustworthy set in accordance with a threshold value of trustworthiness. The attribute weights are adjusted by running a genetic algorithm and are used in the second round of retrieval of the untrustworthy set to obtain the classification results. Experimental results demonstrate that our proposed method performs favorably compared with other methods. Also, the proposed method has less computation complexity for the trustworthiness evaluation, and enhances understanding on thinking and inference for case-based reasoning classifiers. © 2015 Elsevier Ltd. All rights reserved. Source


Li W.,Beijing University of Technology | Li W.,CAS Institute of Automation | Li W.,Beijing Key Laboratory of Computational Intelligence and Intelligent System | Li J.,Capital Medical University | And 6 more authors.
Restorative Neurology and Neuroscience | Year: 2015

Purpose: Previous studies have shown brain reorganizations after early deprivation of auditory sensory. However, changes of grey matter connectivity have not been investigated in prelingually deaf adolescents yet. In the present study, we aimed to investigate changes of grey matter connectivity within and between auditory, language and visual systems in prelingually deaf adolescents. Methods:We recruited 16 prelingually deaf adolescents and 16 age-and gender-matched normal controls, and extracted the grey matter volume as the structural characteristic from 14 regions of interest involved in auditory, language or visual processing to investigate the changes of grey matter connectivity within and between auditory, language and visual systems. Sparse inverse covariance estimation (SICE) was utilized to construct grey matter connectivity between these brain regions. Results: The results show that prelingually deaf adolescents present weaker grey matter connectivity within auditory and visual systems, and connectivity between language and visual systems declined. Notably, significantly increased brain connectivity was found between auditory and visual systems in prelingually deaf adolescents. Conclusions: Our results indicate "cross-modal" plasticity after deprivation of the auditory input in prelingually deaf adolescents, especially between auditory and visual systems. Besides, auditory deprivation and visual deficits might affect the connectivity pattern within language and visual systems in prelingually deaf adolescents. © 2015-IOS Press and the authors. All rights reserved. Source


Li W.,Beijing University of Technology | Li W.,CAS Institute of Automation | Li W.,Beijing Key Laboratory of Computational Intelligence and Intelligent System | Li J.,Capital Medical University | And 5 more authors.
Neural Plasticity | Year: 2016

Previous neuroimaging studies suggested structural or functional brain reorganizations occurred in prelingually deaf subjects. However, little is known about the reorganizations of brain network architectures in prelingually deaf adolescents. The present study aims to investigate alterations of whole-brain functional network using resting-state fMRI and graph theory analysis. We recruited 16 prelingually deaf adolescents (1018 years) and 16 normal controls matched in age and gender. Brain networks were constructed from mean time courses of 90 regions. Widely distributed network was observed in deaf subjects, with increased connectivity between the limbic system and regions involved in visual and language processing, suggesting reinforcement of the processing for the visual and verbal information in deaf adolescents. Decreased connectivity was detected between the visual regions and language regions possibly due to inferior reading or speaking skills in deaf subjects. Using graph theory analysis, we demonstrated small-worldness property did not change in prelingually deaf adolescents relative to normal controls. However, compared with healthy adolescents, eight regions involved in visual, language, and auditory processing were identified as hubs only present in prelingually deaf adolescents. These findings revealed reorganization of brain functional networks occurred in prelingually deaf adolescents to adapt to deficient auditory input. © 2016 Wenjing Li et al. Source


Yang G.,Beijing University of Technology | Yang G.,East China Jiaotong University | Qiao J.F.,Beijing University of Technology | Qiao J.F.,Beijing Key Laboratory of Computational Intelligence and Intelligent System
Applied Soft Computing Journal | Year: 2014

Spatial architecture neural network (SANN), which is inspired by the connecting mode of excitatory pyramidal neurons and inhibitory interneurons of neocortex, is a multilayer artificial neural network and has good learning accuracy and generalization ability when used in real applications. However, the backpropagation-based learning algorithm (named BP-SANN) may be time consumption and slow convergence. In this paper, a new fast and accurate two-phase sequential learning scheme for SANN is hereby introduced to guarantee the network performance. With this new learning approach (named SFSL-SANN), only the weights connecting to output neurons will be trained during the learning process. In the first phase, a least-squares method is applied to estimate the span-output-weight on the basis of the fixed randomly generated initialized weight values. The improved iterative learning algorithm is then used to learn the feedforward-output-weight in the second phase. Detailed effectiveness comparison of SFSL-SANN is done with BP-SANN and other popular neural network approaches on benchmark problems drawn from the classification, regression and time-series prediction applications. The results demonstrate that the SFSL-SANN is faster convergence and time-saving than BP-SANN, and produces better learning accuracy and generalization performance than other approaches. © 2014 Elsevier B.V. All rights reserved. Source


Li F.,Beijing University of Technology | Li F.,University of Jinan | Li F.,Beijing Key Laboratory of Computational Intelligence and Intelligent System | Qiao J.,Beijing University of Technology | And 5 more authors.
Applied Soft Computing Journal | Year: 2016

In this paper, a self-organizing cascade neural network (SCNN) with random weights is proposed for nonlinear system modeling. This SCNN is constructed via simultaneous structure and parameter learning processes. In structure learning, the units, which lead to the maximal error reduction of the network, are selected from the candidates and added to the existing network one by one. A stopping criterion based on the training and validation errors is introduced to select the optimal network size to match with a given application. In parameter learning, the weights connected with the output units are incrementally updated without gradients or generalized inverses, while the other weights are randomly assigned and no need to be tuned. Then, the convergence of SCNN is analyzed. Finally, the proposed SCNN is tested on two benchmark nonlinear systems and an actual municipal sewage treatment system. The experiment results show that the proposed SCNN has better performance on nonlinear system modeling than other similar methods. © 2016 Published by Elsevier B.V. Source

Discover hidden collaborations