| Year |
Technique |
Accuracy |
1992
1995
1996
1998
1999
2002
2004
2005
2006
2012
2012
2013 |
[37] Self-Organizing Feature Map (SOFM)
[57] Integration of Parametric Pattern Recognition algorithm (PPR) and
Artificial Neural Network (ANN)
[1] SOFM and Learning Vector Quantization (LVQ) [16] Modular ANN
[17] SOFM, LVQ and statistical method based on Euclidean distance
[40] Radial Basis Networks (RBN) and Decision Trees
[25] Continuous Wavelet Transform (CWT) and a multi-channel ANN
[27] Multi-Layer Perceptron (MLP)
[76] Wavelet-Based Neural Network (WNN)
[7] SOFM and LVQ
[66] Principal Component Analysis (PCA) and Probabilistic Neural Net- work (PNN)
[2] PCA and PNN |
76-83%
80-90%
60-80%
79.6%.
90%
89%
–
91.6%
90.7%
97.6%.
91.72%
68-94.3% |
2001
2006
2007
2012 |
[14] Fuzzy logic
[63] Adaptive fuzzy k-NN classifier (AFNNC) [35] Pattern Discovery (PD) algorithm
[20] fuzzy logic
|
88.4%
96.6%
–
97% |
1996
2004
2010 |
[56] combined ANN and genetics-based machine learning (GBML) models
[78] Fuzzy integral of multiple ANN [46] Neuro-Fuzzy system (NFS)
|
80%
88.58%
90% |
2002
2005
2009
2010
2010
2010 |
[40] Support Vector Machine (SVM) with one against one training algorithm
[27] SVM
[41] Multiclass SVM
[42] Binary SVM
[42] Fuzzy Support Vector Machine (FSVM) [38] SVM
|
89%
92.3%
100%
100%
99.6%
70.4% |
1995
2012
2012
2013 |
[60] Principle Component Analysis and multivariate discriminant algorithm |
70.4-76.5% |
[44] Decision Tree
[74] FSVM classifier combined with statistical features extracted from
Discrete Wave Transform (DWT)
[75] hybridization of the Particle Swarm Optimization (PSO) and SVM |
96.33-96.50
97.67%
96.75% |
