A Comparison of Supervised Multilayer Back Propagation and Unsupervised Self Organizing Maps for the Diagnosis of Thyroid Disease

Abstract

Artificial Neural Networks have been widely used for the purpose of medical diagnosis in the last decades. The diagnosis of diseases such as thyroid using artificial neural networks is an important research area because of the need of more and more accuracy in the crucial process of disease diagnosis. This paper presents a comparison of two artificial neural network algorithms viz. Multilayer Back Propagation (BPN) -a supervised approach and Self Organizing Maps (SOM) -an unsupervised approach for the diagnosis of thyroid disease using real patient data. It has been found in this study that the results of unsupervised SOM network performed equally well with 100% accuracy as the supervised BPN network in less training time but with a comparatively large percentage of training data.

Exploring the socio-economic and environmental components of infectious diseases using multivariate geovisualization: West Nile Virus

“…The three visualizations used were: (1) a SOM ( Kohonen, 2001 ) to perform multivariate analysis, dimensional reduction, and data reduction; (2) a PCP ( Inselberg, 2002 ) to visualize the multivariate patterns with display; and (3) geographic mapping (GeoMap) to highlight clusters of specific interrelationships. The geovisualization tools of SOM and PCP have been adopted in many fields of science for exploring difficult high dimensional and non-linear problems as well as for visualization of multivariate problems ( Edsall, 2003a , Koua & Kraak, 2004 ; Guo et al, 2005 ; Basara & Yuan, 2008 ; Kaur, Singh & Bahrdwaj, 2013 ; Brookes et al, 2014 ; Fanelli Kuczmarski et al, 2018 ; Mutheneni et al, 2018 ). These tools help to display the high-dimensional datasets, search for hidden relations among the complex set of variables and transform them into a 2-D pattern recognition problem.…”

Peer Review #1 of "Exploring the socio-economic and environmental components of infectious diseases using multivariate geovisualization: West Nile Virus (v0.1)"

“…The three visualizations used were: (1) a self-organizing map (Kohonen 2001) to perform multivariate analysis, dimensional reduction, and data reduction; (2) a parallel coordinate plot (Inselberg 2002) to visualize the multivariate patterns with display; and (3) geographic mapping (GeoMap) to highlight clusters of specific interrelationships. The geovisualization tools of SOM and PCP have been adopted in many fields of science for exploring difficult high dimensional and nonlinear problems as well as for visualization of multivariate problems , Koua and Kraak 2004, Guo, Gahegan et al 2005, Basara and Yuan 2008, Kaur, Singh et al 2013, Brookes, Hernandez-Jover et al 2014, Fanelli Kuczmarski, Bodt et al 2018, Mutheneni, Mopuri et al 2018. These tools help to display the high-dimensional datasets, search for hidden relations among the complex set of variables and transform them into a 2-D pattern recognition problem.…”

Perancangan Pengenalan Karakter Alfabet menggunakan Metode Hybrid Jaringan Syaraf Tiruan

“…Backpropagation merupakan salah satu metode dalam jaringan syaraf tiruan yang menggunakan pembelajaran dengan supervisi yang populer dan memiliki keunggulan dalam kemampuan pembelajarannya. Berdasarkan penelitian Kaur et al (2013) No _page-end_page Backpropagation mencapai 100% akurasi dengan 60% data pelatihan tetapi self organizing maps memerlukan 70% data pelatihan. [1] Oleh karena itu, penulis mencoba melakukan hybrid pada kedua metode tersebut dengan tujuan meningkatkan performansi dan akurasi.…”

“…Berdasarkan penelitian Kaur et al (2013) No _page-end_page Backpropagation mencapai 100% akurasi dengan 60% data pelatihan tetapi self organizing maps memerlukan 70% data pelatihan. [1] Oleh karena itu, penulis mencoba melakukan hybrid pada kedua metode tersebut dengan tujuan meningkatkan performansi dan akurasi.…”