This is an open access article published by the IET under the Creative Commons Attribution -NonCommercial License (http://creativecommons.org/licenses/by-nc/3.0/)
Low-power wearable devices for disease diagnosis are used at anytime and anywhere. These are non-invasive and pain-free for the better quality of life. However, these devices are resource constrained in terms of memory and processing capability. Memory constraint allows these devices to store a limited number of patterns and processing constraint provides delayed response. It is a challenging task to design a robust classification system under above constraints with high accuracy. In this Letter, to resolve this problem, a novel architecture for weightless neural networks (WNNs) has been proposed. It uses variable sized random access memories to optimise the memory usage and a modified binary TRIE data structure for reducing the test time. In addition, a bio-inspired-based genetic algorithm has been employed to improve the accuracy. The proposed architecture is experimented on various disease datasets using its software and hardware realisations. The experimental results prove that the proposed architecture achieves better performance in terms of accuracy, memory saving and test time as compared to standard WNNs. It also outperforms in terms of accuracy as compared to conventional neural network-based classifiers. The proposed architecture is a powerful part of most of the low-power wearable devices for the solution of memory, accuracy and time issues.
References
-
-
1)
-
11. Aleksander, I., Morton, H.: ‘An introduction to neural computing’ (Chapman and Hall London, 1990), vol. 240.
-
2)
-
27. Haykin, S.S.: ‘Neural networks and learning machines’ (Pearson, Upper Saddle River, NJ, USA, 2009), vol. 3.
-
3)
-
24. Forechi, A., De Souza, A.F., de Oliveira Neto, J., et al: ‘Fat-fast VG-RAM WNN: a high performance approach’, Neurocomputing, 2016, 183, pp. 56–69, (doi: 10.1016/j.neucom.2015.06.104).
-
4)
-
5)
-
12. De Gregorio, M.: ‘An intelligent active video surveillance system based on the integration of virtual neural sensors and BDI agents’, IEICE Trans. Inf. Syst., 2008, 91, (7), pp. 1914–1921 (doi: 10.1093/ietisy/e91-d.7.1914).
-
6)
-
26. Fushiki, T.: ‘Estimation of prediction error by using k-fold cross-validation’, Stat. Comput., 2011, 21, (2), pp. 137–146, (doi: 10.1007/s11222-009-9153-8).
-
7)
-
29. Waibel, A., Hanazawa, T., Hinton, G., et al: ‘Phoneme recognition using time delay neural networks’, IEEE Trans. Acoustics Speech Signal Process., 1989, 37, (3), pp. 328–339, (doi: 10.1109/29.21701).
-
8)
-
13. Coraggio, P., De Gregorio, M.: ‘WiSARD and NSP for robot global localization’. Int. Work-Conf. on the Interplay between Natural and Artificial Computation, 2007, pp. 449–458.
-
9)
-
28. Specht, D.F.: ‘Probabilistic neural networks’, Neural Netw., 1990, 3, (1), pp. 109–118, (doi: 10.1016/0893-6080(90)90049-Q).
-
10)
-
23. Cheruku, R., Edla, D.R., Kuppili, V.: ‘SM – Ruleminer: Spider monkey based rule miner using novel fitness function for diabetes classification’, Comput. Biol. Med., 2017, 81, pp. 79–92, (doi: 10.1016/j.compbiomed.2016.12.009).
-
11)
-
12)
-
8. Ludermir, T.B., de Oliveira, W.R.: ‘Weightless neural models’, Comput. Stand. Interfaces, 1994, 16, (3), pp. 253–263, (doi: 10.1016/0920-5489(94)90016-7).
-
13)
-
6. Bledsoe, W.W., Browning, I.: ‘Pattern recognition and reading by machine’ (PGEC, 1959).
-
14)
-
9. Aleksander, I., Thomas, W., Bowden, P.: ‘WiSARD a radical step forward in image recognition’, Sens. Rev., 1984, 4, (3), pp. 120–124 (doi: 10.1108/eb007637).
-
15)
-
14. Do Prado, C.B., Franca, F.M., Costa, E., et al: ‘A new intelligent systems approach to 3d animation in television’. Proc. of the 6th ACM Int. Conf. on Image and Video Retrieval, 2007, pp. 117–119, .
-
16)
-
17)
-
7. Aleksander, I., De Gregorio, M., Franca, F.M.G., et al: ‘A brief introduction to weightless neural systems’. ESANN, Citeseer, 2009.
-
18)
-
18. Austin, J.: ‘RAM-based neural networks’ (World Scientific, 1998), vol. 9.
-
19)
-
16. de Aguiar, E., Forechi, A., Veronese, L., et al: ‘Compressing VG-RAM WNN memory for lightweight applications’. 2014 Int. Joint Conf. on Neural Networks (IJCNN), 2014, pp. 1063–1070, doi: 10.1109/IJCNN.2014.6889563.
-
20)
-
21)
-
15. Prado, C.B., Franca, F.M., Diacovo, R., et al: ‘The influence of order on a large bag of words’. 2008 Eighth Int. Conf. on Intelligent Systems Design and Applications, 2008, vol. 1, pp. 432–436, .
-
22)
-
19. Willard, D.E.: ‘New TRIE data structures which support very fast search operations’, J. Comput. Syst. Sci., 1984, 28, (3), pp. 379–394, (doi: 10.1016/0022-0000(84)90020-5).
-
23)
-
10. Berger, M., Forechi, A., De Souza, A.F., et al: ‘Traffic sign recognition with WiSARD and VG-RAM weightless neural networks’, J. Netw. Innovative Comput., 2013, 1, pp. 87–98.
-
24)
-
25)
-
22. Cheruku, R., Edla, D.R., Kuppili, V.: ‘Diabetes classification using radial basis function network by combining cluster validity index and bat optimization with novel fitness function’, Int. J. Comput. Intell. Syst., 2017, 10, (1), pp. 247–265, (doi: 10.2991/ijcis.2017.10.1.17).
-
26)
-
17. De Souza, A.F., Pedroni, F., Oliveira, E., et al: ‘Automated multilabel text categorization with vgram weightless neural networks’, Neurocomputing, 2009, 72, (10), pp. 2209–2217, (doi: 10.1016/j.neucom.2008.06.028).
-
27)
-
28)
-
29)
-
20. Bodon, F., Ronyai, L.: ‘TRIE: an alternative data structure for data mining algorithms’, Math. Comput. Model., 2003, 38, (7), pp. 739–751, (doi: 10.1016/0895-7177(03)90058-6).
http://iet.metastore.ingenta.com/content/journals/10.1049/htl.2017.0003
Related content
content/journals/10.1049/htl.2017.0003
pub_keyword,iet_inspecKeyword,pub_concept
6
6