Multiclass Alpha Integration of Scores from Multiple Classifiers

Alpha integration methods have been used for integrating stochastic models and fusion in the context of detection (binary classification). Our work proposes separated score integration (SSI), a new method based on alpha integration to perform soft fusion of scores in multiclass classification problems, one of the most common problems in automatic classification. Theoretical derivation is presented to optimize the parameters of this method to achieve the least mean squared error (LMSE) or the minimum probability of error (MPE). The proposed alpha integration method was tested on several sets of simulated and real data. The first set of experiments used synthetic data to replicate a problem of automatic detection and classification of three types of ultrasonic pulses buried in noise (four-class classification). The second set of experiments analyzed two databases (one publicly available and one private) of real polysomnographic records from subjects with sleep disorders. These records were automatically staged in wake, rapid eye movement (REM) sleep, and non-REM sleep (three-class classification). Finally, the third set of experiments was performed on a publicly available database of single-channel real electroencephalographic data that included epileptic patients and healthy controls in five conditions (five-class classification). In all cases, alpha integration performed better than the considered single classifiers and classical fusion techniques.

[1]  B. Hjorth.  The physical significance of time domain descriptors in EEG analysis. , 1973, Electroencephalography and clinical neurophysiology.

[2]  Jiri Matas,et al.  On Combining Classifiers , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  K Lehnertz,et al.  Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: dependence on recording region and brain state. , 2001, Physical review. E, Statistical, nonlinear, and soft matter physics.

[4]  Samy Bengio,et al.  How do correlation and variance of base-experts affect fusion in biometric authentication tasks? , 2005, IEEE Transactions on Signal Processing.

[5]  Torbjørn Eltoft,et al.  Modeling the amplitude statistics of ultrasonic images , 2006, IEEE Transactions on Medical Imaging.

[6]  S. Amari.  Integration of Stochastic Models by Minimizing -Divergence , 2007, Neural Computation.

[7]  Dalei Wu.  Parameter Estimation for -GMM Based on Maximum Likelihood Criterion , 2009, Neural Computation.

[8]  Yoonsuck Choe,et al.  Learning α-integration with partially-labeled data , 2010, 2010 IEEE International Conference on Acoustics, Speech and Signal Processing.

[9]  Addisson Salazar,et al.  ICA Mixtures Applied to Ultrasonic Nondestructive Classification of Archaeological Ceramics , 2010, EURASIP J. Adv. Signal Process..

[10]  Ronald M. Summers,et al.  Fusion of machine intelligence and human intelligence for colonic polyp detection in CT colonography , 2011, 2011 IEEE International Symposium on Biomedical Imaging: From Nano to Macro.

[11]  Hlaing Minn,et al.  Real-Time Sleep Apnea Detection by Classifier Combination , 2012, IEEE Transactions on Information Technology in Biomedicine.

[12]  Joseph N. Wilson,et al.  Twenty Years of Mixture of Experts , 2012, IEEE Transactions on Neural Networks and Learning Systems.

[13]  Fakhri Karray,et al.  Multisensor data fusion: A review of the state-of-the-art , 2013, Inf. Fusion.

[14]  Yoonsuck Choe,et al.  Parameter Learning for Alpha Integration , 2013, Neural Computation.

[15]  Joaquín Abellán,et al.  Improving experimental studies about ensembles of classifiers for bankruptcy prediction and credit scoring , 2014, Expert Syst. Appl..

[16]  Mohamed Moshrefi-Torbati,et al.  Signal processing techniques applied to human sleep EEG signals - A review , 2014, Biomed. Signal Process. Control..

[17]  Jonathon A. Chambers,et al.  Audiovisual Speech Source Separation: An overview of key methodologies , 2014, IEEE Signal Processing Magazine.

[18]  João Paulo Papa,et al.  EEG signal classification for epilepsy diagnosis via optimum path forest - A systematic assessment , 2014, Neurocomputing.

[19]  Christian Jutten,et al.  Multimodal Data Fusion: An Overview of Methods, Challenges, and Prospects , 2015, Proceedings of the IEEE.

[20]  Mohamed Abdel Fattah,et al.  New term weighting schemes with combination of multiple classifiers for sentiment analysis , 2015, Neurocomputing.

[21]  Antonio Soriano,et al.  Fusion of Scores in a Detection Context Based on Alpha Integration , 2015, Neural Computation.

[22]  Shun-ichi Amari,et al.  Information Geometry and Its Applications , 2016 .

[23]  Fuqiang Chen,et al.  Automatic sleep stage classification based on sparse deep belief net and combination of multiple classifiers , 2016 .

[24]  Antonio Soriano,et al.  On the fusion of non-independent detectors , 2016, Digit. Signal Process..

[25]  Jasmin Kevric,et al.  An effective combining classifier approach using tree algorithms for network intrusion detection , 2017, Neural Computing and Applications.

[26]  Gang Hua,et al.  A Cycle Deep Belief Network Model for Multivariate Time Series Classification , 2017 .

[27]  Mohamed Deriche,et al.  Classifiers Combination Techniques: A Comprehensive Review , 2018, IEEE Access.

[28]  Jolanda Palmisani,et al.  Pattern Recognition and Anomaly Detection by Self-Organizing Maps in a Multi Month E-nose Survey at an Industrial Site , 2020, Sensors.