Jiancheng Sun

Publications (14)5.57 Total impact

• Article: Scaling the kernel function based on the separating boundary in input space: A data-dependent way for improving the performance of kernel methods.

  Jiancheng Sun, Xiaohe Li, Yong Yang, Jianguo Luo, Yaohui Bai
  Inf. Sci. 01/2012; 184:140-154.
• Article: Denoised P300 and machine learning-based concealed information test method.

  Junfeng Gao, Xiangguo Yan, Jiancheng Sun, Chongxun Zheng
  [show abstract] [hide abstract]
  ABSTRACT: In this paper, a novel P300-based concealed information test (CIT) method was proposed to improve the efficiency of differentiating deception and truth-telling. Thirty subjects including the guilty and innocent performed the paradigm based on three types of stimuli. In order to reduce the influence from the occasional variability of cognitive states on the CIT, several single-trials from Pz in probe stimuli within each subject were first averaged. Then the three groups of features were extracted from these averaged single-trials. Finally, two classes of feature samples were used to train a support vector machine (SVM) classifier. Meanwhile, the optimal number of averaged Pz waveforms and some other parameter values in the classifiers were determined by the cross validation procedures. Results show that if choosing accuracy of 90% as a detecting standard of P3 component to classify a subject's status (guilty or innocent), our method can achieve individual diagnostic rate of 100%. The individual diagnostic rate of our method was higher than the results of the other related reports. The presented method improves efficiency of CIT, and is more practical, lower fatigue and less countermeasure behavior in comparison with previous report methods, which could extend the laboratory study to the practical application.
  Computer methods and programs in biomedicine 11/2010; 104(3):410-7. · 1.14 Impact Factor
• Article: Automatic removal of various artifacts from EEG signals using combined methods.

  Junfeng Gao, Yong Yang, Jiancheng Sun, Gang Yu
  [show abstract] [hide abstract]
  ABSTRACT: In this article, a novel and robust method is proposed to automatically remove various artifacts from EEG signals. First, canonical correlation analysis method is adopted to separate electromyography (EMG) artifacts from EEG signals. EMG-free EEG signals are obtained by subtracting the contribution of the components with autocorrelation value less than a threshold determined by the statistical analysis. For the removal of ocular artifacts, independent component analysis is applied to decompose the EMG-free signals. For the identification of eye movement artifact components, spectral and topographic features are extracted, and the classifier of support vector machine is used. Specifically, a peak detection algorithm of independent component is proposed to identify eye blink artifact components for the first time. The proposed artifact removal method is evaluated by the comparisons of EEG data before and after artifacts removal. The results show that the proposed method provides a promising method for complete artifact removal from EEG.
  Journal of clinical neurophysiology: official publication of the American Electroencephalographic Society 10/2010; 27(5):312-20. · 1.47 Impact Factor
• Article: Analysis of the distance between two classes for tuning SVM hyperparameters.

  Jiancheng Sun, Chongxun Zheng, Xiaohe Li, Yatong Zhou
  [show abstract] [hide abstract]
  ABSTRACT: An important step in the construction of a support vector machine (SVM) is to select optimal hyperparameters. This paper proposes a novel method for tuning the hyperparameters by maximizing the distance between two classes (DBTC) in the feature space. With a normalized kernel function, we find that DBTC can be used as a class separability criterion since the between-class separation and the within-class data distribution are implicitly taken into account. Employing DBTC as an objective function, we develop a gradient-based algorithm to search the optimal kernel parameter. On the basis of the geometric analysis and simulation results, we find that the optimal algorithm and the initialization problem become very simple. Experimental results on the synthetic and real-world data show that the proposed method consistently outperforms other existing hyperparameter tuning methods.
  IEEE Transactions on Neural Networks 02/2010; 21(2):305-18. · 2.95 Impact Factor
• Article: Distance between Two Classes: A Novel Kernel Class Separability Criterion.

  Jiancheng Sun, Chongxun Zheng, Xiaohe Li
  IEICE Transactions. 01/2009; 92-D:1397-1400.
• Conference Proceeding: A template-based isomap algorithm for real-time removal of ocular artifacts from EEG signals.

  Junfeng Gao, Jiancheng Sun, Pan Lin, Chongxun Zheng, Pei Wang
  Proceedings of the 2nd International Conference on Interaction Sciences: Information Technology, Culture and Human 2009, Seoul, Korea, 24-26 November 2009; 01/2009
• Article: Noise reduction for fast fading channel by Recurrent Least Squares Support Vector Machines in embedding phase spaces

  Zheng Xiang, Taiyi Zhang, Jiancheng Sun
  [show abstract] [hide abstract]
  ABSTRACT: A new strategy for noise reduction of fast fading channel is presented. Firstly, more information is acquired utilizing the reconstructed embedding phase space. Then, based on the Recurrent Least Squares Support Vector Machines (RLS-SVM), noise reduction of the fast fading channel is realized. This filtering technique does not make use of the spectral contents of the signal. Based on the stability and the fractal of the chaotic attractor, the RLS-SVM algorithm is a better candidate for the nonlinear time series noise-reduction. The simulation results shows that better noise-reduction performance is acquired when the signal to noise ratio is 12dB.
  Journal of Electronics (China) 10/2006; 23(6):926-928.
• Conference Proceeding: Noise Reduction of Chaotic Systems Based on Least Squares Support Vector Machines

  Jiancheng Sun, Yatong Zhou
  [show abstract] [hide abstract]
  ABSTRACT: In order to resolve the noise reduction in chaotic system, a novel idea based on least square support vector machine (LS-SVM) is proposed in this paper. By analyzing the relationship between the function approximation and the noise reduction, we realized that the noise reduction can be implemented by the function approximation techniques. On the basis of the LS-SVM, the function approximation is carried out and the noise reduction achieved simultaneously
  Communications, Circuits and Systems Proceedings, 2006 International Conference on; 07/2006
• Conference Proceeding: Music Style Classification with a Novel Bayesian Model.

  Yatong Zhou, Taiyi Zhang, Jiancheng Sun
  Advanced Data Mining and Applications, Second International Conference, ADMA 2006, Xi'an, China, August 14-16, 2006, Proceedings; 01/2006
• Conference Proceeding: Nonlinear Noise Reduction of Chaotic Time Series Based on Multi-dimensional Recurrent Least Squares Support Vector Machines.

  Jiancheng Sun, Yatong Zhou, Yaohui Bai, Jianguo Luo
  Neural Information Processing, 13th International Conference, ICONIP 2006, Hong Kong, China, October 3-6, 2006, Proceedings, Part I; 01/2006
• Conference Proceeding: Modelling of Chaotic Systems with Recurrent Least Squares Support Vector Machines Combined with Reconstructed Embedding Phase Space.

  Zheng Xiang, Taiyi Zhang, Jiancheng Sun
  Advances in Natural Computation, First International Conference, ICNC 2005, Changsha, China, August 27-29, 2005, Proceedings, Part I; 01/2005
• Conference Proceeding: Nonlinear prediction of mobile-radio fading channel using recurrent least squares support vector machines and embedding phase space

  Jiancheng Sun, Taiyi Zhang, Feng Liu
  [show abstract] [hide abstract]
  ABSTRACT: Prediction of the rapidly fading mobile-radio channel enables a number of capacity improving techniques, such as fast resource allocation or fast adaptive modulation. We construct an embedding phase space which includes more system information than the scalar time series; then we use a new nonlinear regression method, recurrent least squares support vector machines (RLS-SVM), to resolve the prediction problem. A performance evaluation of the prediction algorithm is carried out with various SNR values on Rayleigh fading channels. The simulation results show that the proposed algorithm is a good method for long range prediction of the fading channel.
  Communications, Circuits and Systems, 2004. ICCCAS 2004. 2004 International Conference on; 07/2004
• Conference Proceeding: Modelling of Chaotic Systems with Novel Weighted Recurrent Least Squares Support Vector Machines.

  Jiancheng Sun, Taiyi Zhang, Haiyuan Liu
  Advances in Neural Networks - ISNN 2004, International Symposium on Neural Networks, Dalian, China, August 19-21, 2004, Proceedings, Part I; 01/2004
• Conference Proceeding: Nonlinear prediction of fast fading channel parameters based on the chaotic attractor

  Jiancheng Sun, Taiyi Zhang, Feng Liu
  [show abstract] [hide abstract]
  ABSTRACT: Prediction of the rapidly fading mobile-radio channel enables a number of capacity improving techniques such as the fast resource allocation or fast adaptive modulation. In this paper, we acquire the chaotic attractor by reconstructing the embedding phase space which include more information of system than the scalar time series, and then prediction is carried out in the embedding phase space by the select neighbors of the prediction point with a weighted average of the neighbors. A performance evaluation of the prediction algorithm is analyzed with various SNR on Rayleigh fading channels. The simulation result show that the proposed algorithm is a good candidate for long range prediction of fading channel.
  Emerging Technologies: Frontiers of Mobile and Wireless Communication, 2004. Proceedings of the IEEE 6th Circuits and Systems Symposium on;