A Predictive Analysis of Heart Diseases and Diabetes Using Adaptive Modified Backpropagation

 
 
 
  • Abstract
  • Keywords
  • References
  • PDF
  • Abstract


    An Artificial Neural Network (ANN) is a data preparing method that is motivated by the way organic sensory systems, for example, the mind, process data. ANN learns (i.e. logically enhance execution on) errands by thinking about illustrations, for the most part without assignment particular programming. Backpropagation is a method utilized as a part of neural networks to figure a value that is required in the computation of the weights to be utilized as a part of the system. In this project, neural network Backpropagation with adaptive learning rate is used to predict heart and diabetes diseases to the patients by the given parameters. The algorithm modifies the learning rate at each step based on the data and gives faster results when compared to the traditional algorithm. This is chosen to implement on Hear Diseases and Diabetic Disease dataset and results are compared. The model is trained with a set of training data and then that trained model is used to predict the testing data of patient disease. These predictions will give an idea of the type of disease and can be used to diagnose the patient.

     

     


  • Keywords


    Artificial neural networks, feed forward neural networks, backpropagation, adaptive learning rate, heart disease and diabetic disease datasets.

  • References


      [1] Kriesel D, A Brief Introduction to Neural Networks, dkriesel.com, (2009).

      [2] Kumar P & Sharma P, “Artificial Neural Networks-A Study”, International Journal of Emerging Engineering Research and Technology, Vol.2, No.2,(2014), pp.143-148.

      [3] Jeyaseeli Subavathia S & Kathirvalavakumar T, “Adaptively modified backpropagation algorithm-based on differential errors”, International Journal of Computer Science, Engineering and Applications (IJCSEA), Vol.1, No.5, (2011).

      [4] Maind SB & Wankar P, “Research paper on basic of artificial neural network”, International Journal on Recent and Innovation Trends in Computing and Communication, Vol.2, No.1,(2014), pp.96-100.

      [5] Cilimkovic M, “Neural networks and back propagation algorithm”, Institute of Technology Blanchardstown, Blanchardstown Road North Dublin, (2015).

      [6] Izawati N & Alias B, “Effects Of Normalization And Training Algorithms On MLP Network Performances”.

      [7] Rotich N, “Forecasting of wind speeds and directions with artificial neural networks”, (2014).

      [8] https://mattmazur.com/2015/03/17/a-step-by-step-backpropagation-example/

      [9] http://playground.tensorflow.org

      [10] https://visualstudiomagazine.com/articles/2014/01/01/how-to-standardize-data-for-neural-networks.aspx

      [11] https://visualstudiomagazine.com/articles/2013/07/01/neural-network-data-normalization-and-encoding.aspx

      [12] https://www.researchgate.net/post/How_to_decide_the_number_of_hidden_layers_and_nodes_in_a_hidden_layer

      [13] https://www.solver.com/training-artificial-neural-network-intro

      [14] https://www.neuraldesigner.com/blog/5_algorithms_to_train_a_neural_network

      https://www.analyticsvidhya.com/blog/2017/10/fundamentals-deep-learning-activation-functions-when-to-use-them/

 

View

Download

Article ID: 15131
 
DOI: 10.14419/ijet.v7i3.6.15131




Copyright © 2012-2015 Science Publishing Corporation Inc. All rights reserved.