Probabilistic neural networks (PNNs) are artificial neural network algorithms widely used in pattern recognition and classification problems. In the traditional PNN algorithm, the probability density function (PDF) is approximated using the entire training dataset for each class. In some complex datasets, classmate clusters may be located far from each other and these distances between clusters may cause a reduction in the correct class's posterior probability and lead to misclassification. This paper presents a novel PNN algorithm, the competitive probabilistic neural network (CPNN). In the CPNN, a competitive layer ranks kernels for each class and an optimum fraction of kernels are selected to estimate the class-conditional probability. Using a stratified, repeated, random subsampling cross-validation procedure and 9 benchmark classification datasets, CPNN is compared to both traditional PNN and the state of the art(e.g. enhanced probabilistic neural network, EPNN). These datasets are examined with and without noise and the algorithm is evaluated with several ratios of training to testing data. In all datasets (225 simulation categories), performance percentages of both CPNN and EPNN are greater than or equivalent to that of the traditional PNN; in 73% of simulation categories, the CPNN analyses show modest improvement in performance over the state of the art.
AdeliH. and KarimA., Fuzzy-wavelet RBFNN model for freeway incident detection, Journal of Transportation Engineering126(6) (2000), 464-471. doi: 10.1061/(ASCE)0733-947X(2000)126:6(464).
2.
AdeliH. and PanakkatA., A probabilistic neural network for earthquake magnitude prediction, Neural Networks22(7) (2009), 1018-1024. doi: 10.1016/j.neunet.2009.05.003.
3.
AdeliH. and ParkH.S., Counterpropagation neural networks in structural engineering, Journal of Structural Engineering121(8) (1995), 1205-1212. doi: 10.1061/(ASCE)0733-9445(1995)121:8(1205).
4.
AhmadlouM. and AdeliH., Enhanced probabilistic neural network with local decision circles: A robust classifier, Integrated Computer-Aided Engineering17(3) (2010), 197-210. doi: 10.3233/ICA-2010-0345.
5.
AkramM. and LotfifardS., Modeling and health monitoring of DC side of photovoltaic array, IEEE Transactions on Sustainable Energy6(4) (2015), 1245-1253. doi: 10.1109/TSTE.2015.2425791.
6.
AlweshahM. and AbdullahS., Hybridizing firefly algorithms with a probabilistic neural network for solving classification problems, Applied Soft Computing Journal35 (2015), 513-524. doi: 10.1016/j.asoc.2015.06.018.
7.
AzarA. and El-SaidS., Probabilistic neural network for breast cancer classification, Neural Computing and Applications23(6) (2012), 1737-1751. doi: 10.1007/s00521-012-1134-8.
8.
BabichG. and CampsO., Weighted Parzen windows for pattern classification, IEEE Transactions on Pattern Analysis and Machine Intelligence18(5) (1996), 567-574. doi: 10.1109/34.494647.
9.
AliJ.B., SaidiL., MouelhiA., Chebel-MorelloB. and FnaiechF., Linear feature selection and classification using PNN and SFAM neural networks for a nearly online diagnosis of bearing naturally progressing degradations, Engineering Applications of Artificial Intelligence42 (2015), 67-81. doi: 10.1016/j.engappai.2015.03.013.
10.
BishopC., Neural networks for pattern recognition, United States. Oxford University Press Inc; 2000.
11.
BurrascanoP., Learning vector quantization for the probabilistic neural network, IEEE Transactions on Neural Networks2(4) (1991), 458-461. doi: 10.1109/72.88165.
12.
CacoullosT., Estimation of a multivariate density, Annals of the Institute of Statistical Mathematics18(1) (1966), 179-189.
13.
ChabukT., ReggiaJ., LohnJ. and LindenD., Causally-guided evolutionary optimization and its application to antenna array design, Integrated Computer Aided Engineering19(2) (2012), 111-124. doi: 10.3233/ICA-2012-0395.
14.
ColettaL., HruschkaE., AcharyaA. and GhoshJ., Using metaheuristics to optimize the combination of classifier and cluster ensembles, Integrated Computer Aided Engineering22(3) (2015), 229-242. doi: 10.3233/ICA-150485.
15.
CortesC. and VapnikV., Support-vector networks, Machine Learning20(3) (1995), 273-297.
16.
DuinR., On the choice of smoothing parameters for Parzen estimators of probability density functions, IEEE Transactions on ComputersC-25(11) (1976), 1175-1179.
17.
EgorovA. and MikshinaV., Use of probabilistic neural networks to solve classification problems in surgery, Optical Memory and Neural Networks (Information Optics)20(3) (2011), 224-231. doi: 10.3103/S1060992X11030015.
18.
FaslJ., HelwigT. and WoodS., Probabilistic method for estimating remaining fatigue life in steel bridges using measured strain data, Journal of Civil Structural Health Monitoring3(4) (2013), 317-324. doi: 10.1007/s13349-013-0056-1.
19.
FrankA. and AsuncionA., 2010 UCI machine learning repository http://archive.ics.uci.edu/ml. University of California, School of Information and Computer Science, Irvine, CA, USA.
20.
GeorgiouV., AlevizosP. and VrahatisM., Novel approaches to probabilistic neural networks through bagging and evolutionary estimating of prior probabilities, Neural Processing Letters27(2) (2008), 153-162. doi: 10.1007/s11063-007-9066-5.
21.
GeorgiouV., PavlidisN., ParsopoulosK., AlevizosP. and VrahatisM., New self-adaptive probabilistic neural networks in bioinformatic and medical tasks, International Journal on Artificial Intelligence Tools15(3) (2006), 371-396. doi: 10.1142/S0218213006002722.
22.
GeorgiouV. and VrahatisM., Fuzzy evolutionary probabilistic neural networks, in: IAPR Workshop on Artificial Neural Networks in Pattern Recognition, PrevostL. et al., eds, Springer Berlin Heidelberg, Paris, France 1 (2008), 113-124.
23.
GerbecD., GašpericS., ŘmonI. and GubinaF., Allocation of the load profiles to consumers using probabilistic neural networks, IEEE Transactions on Power Systems20(2) (2005), 548-555. doi: 10.1109/TPWRS.2005.846236.
24.
Ghosh-DastidarS. and AdeliH., Spiking neural networks, International Journal of Neural Systems19(4) (2009), 295-308. doi: 10.1142/S0129065709002002.
25.
Ghosh-DastidarS., AdeliH. and DedmehrN., Principal component analysis-enhanced cosine radial basis function neural network for robust epilepsy and seizure detection, IEEE Transactions on Biomedical Engineering55(2) (2008), 512-518. doi: 10.1109/TBME.2007.905490.
26.
GoodwinM. and YazidiA., A pattern recognition approach for peak prediction of electrical consumption, Integrated Computer-Aided Engineering23(2) (2016), 101-113. doi: 10.1007/978-3-662-44654-6_26.
27.
HartiganJ., Bayes theory, Springer Series in statics, New York, United States, Springer-Verlag, 1983.
28.
HaykinS., Neural networks a comprehensive foundation, United States, Macmillan College Publishing Company Inc, 1994.
29.
HirschauerT., AdeliH. and BufordJ., Computer-aided diagnosis of Parkinson's disease using enhanced probabilistic neural network, Journal of Medical Systems39(11) (2015), 179-190. doi: 10.1007/s10916-015-0353-9.
30.
HoyaT., On the capability of accommodating new classes within probabilistic neural networks, IEEE Transactions on Neural Networks14(2) (2003), 450-453. doi: 10.1109/TNN.2003.809417.
31.
HuangC. and LiaoW., Application of probabilistic neural networks to the class prediction of leukemia and embryonal tumor of central nervous system, Neural Processing Letter19(3) (2004), 211-226. doi: 10.1023/B:NEPL.0000035613.51734.48.
32.
HungS. and AdeliH., A parallel genetic/neural network learning algorithm for MIMD shared memory machines, IEEE Transactions on Neural Networks5(6) (1994), 900-909. doi: 10.1109/72.329686.
33.
HylandD. and FryG., A neural-genetic hybrid approach for optimizing structural health monitoring systems, in: Proceedings of the 2nd International Workshop on Structural Health MonitoringChangF., ed., Technomic Publishing Co, Inc, Stanford, CA, USA 12000, pp. 800-811.
34.
IounousseJ., Er-RakiS., El MotassadeqA. and ChehouaniH., Using an unsupervised approach of Probabilistic Neural Network (PNN) for land use classification from multitemporal satellite images, Applied Soft Computing Journal30 (2015), 1-13. doi: 10.1016/j.asoc.2015.01.037.
35.
JiaL., WangY. and FanL., Multiobjective bilevel optimization for production-distribution planning problems using hybrid genetic algorithm, Integrated Computer-Aided Engineering21(1) (2014), 77-90. doi: 10.3233/ICA-130452.
36.
JiangX. and AdeliH., Neuro-genetic algorithm for non-linear active control of structures, International Journal for Numerical Methods in Engineering75(7) (2008), 770-786. doi: 10.1002/nme.2274.
37.
KarimA. and AdeliH., Radial basis function neural network for work zone capacity and queue estimation, Journal of Transportation Engineering129(5) (2003), 494-503. doi: 10.1061/(ASCE)0733-947X(2003)129:5(494).
38.
KhasheiM. and BijariM., Hybridization of the probabilistic neural networks with feed-forward neural networks for forecasting, Engineering Applications of Artificial Intelligence25(6) (2012), 1277-1288. doi: 10.1016/j.engappai.2012.01.019.
39.
KhasheiM., BijariM. and ArdaliG.R., Hybridization of autoregressive integrated moving average (ARIMA) with probabilistic neural networks (PNNs), Computers & Industrial Engineering63(1) (2012), 37-45. doi: 10.1016/j.cie.2012.01.017.
40.
KumarH. and SrinivasanS., Classification of ovary abnormality using the probabilistic neural network (PNN), Technology and Health Care22(6) (2014), 857-865. doi: 10.3233/THC-140863.
41.
KusyM. and ZajdelR., Application of reinforcement learning algorithms for the adaptive computation of the smoothing parameter for probabilistic neural network, IEEE Transactions on Neural Networks and Learning Systems26(9) (2014), 2163-2175. doi: 10.1109/TNNLS.2014.2376703.
42.
Lee-JyunC., Damage localization for bridges using probabilistic neural networks, KSCE Journal of Civil Engineering11(2) (2016), 111-120. doi: 10.1007/BF02823854.
43.
LiP., Structural damage localization using probabilistic neural networks, Mathematical and Computer Modelling54(3-4) (2011), 965-969. doi: 10.1016/j.mcm.2010.11.023.
44.
LiX., LiW., CaiX. and HeF., A hybrid optimization approach for sustainable process planning and scheduling, Integrated Computer-Aided Engineering22(4) (2015), 311-326. doi: 10.3233/ICA-150492.
45.
LiX., ZhangS., LiS. and ChenJ., An improved method of speech recognition based on probabilistic neural network ensembles, in: 11th International Conference on Natural Computation (ICNC) IEEE, ChangSha, China 1 (2015), 650-654.
46.
LotfiA. and BenyettouA., Over-fitting avoidance in probabilistic neural networks, in: Proceedings of (WCITCA) 2015 World Congress on Information Technology and Computer Applications IEEE, Hammamet, Tunisia 1 (2015), 1-6.
47.
LuM., AbouRizkS. and HermannU., Estimating labor productivity using probability inference neural network, Journal of Computing in Civil Engineering14(4) (2000), 241-248. doi: 10.1061/(ASCE)0887-3801(2000)14:4(241).
48.
ManoukianE., Modern concepts and theorems of mathematical statistics, Springer series in statics, New York, United States. Springer-Verlag, 1985.
49.
MaoK. and TanK., Probabilistic neural-network structure determination for pattern classification, IEEE Transactions on Neural Networks11(4) (2000), 1009-1016. doi: 10.1109/72.857781.
50.
MenhajM. and DelgoshaF., A soft probabilistic neural network for implementation of Bayesian classifiers, in: Proceedings of International Joint Conference on Neural Networks IJCNN'01 IEEE, Washington, DC, USA 1 (2001), 454-458.
51.
OzyildirimB. and AvciM., One pass learning for generalized classifier neural network, Neural Networks73 (2016), 70-76. doi: 10.1016/j.neunet.2015.10.008.
52.
PandeA. and Abdel-atyM., A computing approach using probabilistic neural networks for instantaneous appraisal of rear-end crash risk, Computer-Aided Civil and Infrastructure Engineering23(7) (2008), 549-559. doi: 10.1111/j.1467-8667.2008.00559.x.
53.
ParzenE., On estimation of a probability density function and mode, Annals of Mathematical Statistics33(3) (1962), 1065-1076.
54.
QuanJ., WenX. and XuX., Multiscale probabilistic neural network method for SAR image segmentation, Applied Mathematics and Computation205(2) (2008), 578-583. doi: 10.1016/j.amc.2008.05.030.
55.
RaghuP. and YegnanarayanaB., Supervised texture classification using a probabilistic neural network and constraint satisfaction model, IEEE Transactions on Neural Network9(3) (1998), 516-522. doi: 10.1109/72.668893
56.
RamakrishnanS. and SelvanS., Image texture classification using wavelet packet transform and probabilistic neural network, Intelligent Data Analysis11(6) (2007), 691-705.
57.
RigosA., TsekourasG., VousdoukasM., ChatzipavlisA. and VelegrakisA., A Chebyshev polynomial radial basis function neural network for automated shoreline extraction from coastal imagery, Integrated Computer-Aided Engineering (2016), 1-20. doi: 10.3233/ICA-150507.
58.
RumelhartD., HintonG. and WilliamsR., Learning internal representations by error propagation (No. ICS-8506) Ch. 8. California University, San Diego LA, United States, Jolla Institute for Cognitive Science, 1985.
59.
SamantarayS., PanigrahiB. and DashP., High impedance fault detection in power distribution networks using time-frequency transform and probabilistic neural network, IET Generation, Transmission and Distribution2(2) (2008), 261-270. doi: 10.1049/iet-gtd:20070319.
60.
SavchenkoA., Pattern recognition and increasing of the computational efficiency of a parallel realization of the probabilistic neural network with homogeneity testing, Optical Memory and Neural Networks (Information Optics)22(3) (2013), 184-192. doi: 10.3103/S1060992X13030090.
61.
SavchenkoA., Probabilistic neural network with homogeneity testing in recognition of discrete patterns set, Neural Networks46 (2013), 227-241. doi: 10.1016/j.neunet.2013.06.003.
62.
SittonJ. and StoryB., Estimating soil classification via quantitative and qualitative field testing for use in constructing compressed earth blocks, Procedia Engineering145 (2016), 860-867. doi: 10.1016/j.proeng.2016.04.112.
63.
SpechtD., Generation of polynomial discriminant functions for pattern recognition, PhD Dissertation, Stanford University, 1966.
64.
SpechtD., Generation of polynomial discriminant functions for pattern recognition, IEEE Transactions on Electronic Computers16(3) (1967), 308-319. doi: 10.1109/PGEC.1967.264667.
65.
SpechtD., Probabilistic neural networks for classification, mapping, or associative memory, in: IEEE International Conference on Neural Networks IEEE, USA 1 (1988), 525-532.
StoryB., Structural impairment detection using arrays of competitive artificial neural networks, PhD Dissertation, Texas A&M University, 2012.
68.
StoryB. and FryG., A competitive array of artificial neural networks for use in structural impairment detection, in: Proceedings of the 9th International Workshop on Structural Health MonitoringChangF., ed., DEStech Publications, Stanford, CA, USA 2 (2013), pp. 1753-1760.
69.
StoryB. and FryG., A structural impairment detection system using competitive arrays of artificial neural networks, Computer-Aided Civil and Infrastructure Engineering29(3) (2014), 180-190. doi: 10.1111/mice.12040.
70.
StoryB. and FryG., Methodology for designing diagnostic data streams for use in a structural impairment detection system, Journal of Bridge Engineering19(4) (2014), 04013020. doi: 10.1061/(ASCE)BE.1943-5592.0000556.
71.
TaoH., ZainJ., AhmedM., AbdallaA. and JingW., A wavelet-based particle swarm optimization algorithm for digital image watermarking, Integrated Computer-Aided Engineering19(1) (2012), 81-91. doi: 10.3233/ICA-2012-0392.
72.
TongY., The multivariate normal distribution, Springer series in statics, New York, United States, Springer-Verlag, 1990.
73.
TorfsP. and WojcikR., Local probabilistic neural networks in hydrology, Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere26(1) (2001), 9-14. doi: 10.1016/S1464-1909(01)85006-1.
74.
TranH., PereraB. and NgA., Predicting structural deterioration condition of individual storm-water pipes using probabilistic neural networks and multiple logistic regression models, Journal of Water Resources Planning and Management135(6) (2009), 553-557. doi: 10.1061/(ASCE)0733-9496(2009)135:6(553).
75.
TravenH., A neural network approach to statistical pattern classification by semiparametric estimation of probability density functions, IEEE Transactions on Neural Networks2(3) (1991), 366-377. doi: 10.1109/72.97913.
76.
TripathyM., MaheshwariR. and VermaH., Power transformer differential protection based on optimal probabilistic neural network, IEEE Transactions on Power Delivery25(1) (2010), 102-112. doi: 10.1109/TPWRD.2009.2028800.
77.
ÜbeyliE., Probabilistic neural networks employing Lyapunov exponents for analysis of Doppler ultrasound signals, Computers in Biology and Medicine38(1) (2008), 82-89. doi: 10.1016/j.compbiomed.2007.07.004.
YuY. and McKelveyT., A robust subspace classification scheme based on empirical intersection removal and sparse approximation, Integrated Computer-Aided Engineering22(1) (2015), 59-69. doi: 10.3233/ICA-140470.
80.
ZeinaliY. and StoryB., Structural impairment detection using deep counter propagation neural networks, Procedia Engineering145 (2016), 868-875. doi: 10.1016/j.proeng.2016.04.113.
81.
ZhaiJ. and ZhaoW., Ensemble of multiresolution probabilistic neural network classifiers with fuzzy integral for face recognition, Journal of Intelligent & Fuzzy Systems31(1) (2016), 405-414. doi: 10.3233/IFS-162153.
82.
ZhangC., WangH. and WuM., EEG-based expert system using complexity measures and probability density function control in alpha sub-band, Integrated Computer-Aided Engineering20(4) (2013), 391-405. doi: 10.3233/ICA-130439.
