|
|
August 2022 Issue Vol.12 No.8
Performance Analysis of Multilayer Perceptron in Lung Carcinoma Detection with Google, Respiratory and LC Datasets
https://ia904700.us.archive.org/11/items/ijitce-august-2022/vol12no801.pdf
Dr. S . Karthigai
Assistant Professor and Head in Computer Science,
Dr.R.A.N.M. Arts and Science College,
Erode , Tamil Nadu, India.
Dr. S. Prasath
Assistant Professor, School of Computer Science,
VET Institute of Arts and Science (Co-Edu) College, Erode, Tamil Nadu, India.
Abstract:
An artificial neural network is a mathematical model that simulates the structure and function of the interconnected
neurons in the hidden layer. The neural network makes prediction of training and test set in the input layer which is discussed in the
chapter four. ANN has three layers. They are the input, hidden and the output Layers are applied for different datasets. The intermediate
layer or hidden layer consists of number of neurons. The input and output consists of the single layer. The intermediate layer is
considered as an engine of the whole network which deals with the non linear activation function and sensational domination in the
result. The number of neurons determines the quality of the network in all the three layers for google, Respiratory and LC data. LC dataset is
applied in the input layer and it provides better result which is sent to the intermediate layer of MLP. The stage of patient’s role plays an
important play in improving the three layers of MLP. The three layers are input, intermediate and output layer. Each layer classifies the LC
stages from input to the output layer. The output layer is the final result of the LC patient. The patient’s stage is decided by applying MLP
which is executed in WEGA. The MLP is present all the Artificial Neural Network.
Keywords: Multi Layer Perceptron, Artificial Neural Network, Iterative K Fold Method, Standard Approach, Multi logit regression, Maximum A
Posteriori, Hyperbolic cosine loss function.
