Controlling the Behavior of a Neural Network Weights Using Variables Correlation and Posterior Probabilities Estimation

Journal Title: IOSR Journals (IOSR Journal of Computer Engineering) - Year 2014, Vol 16, Issue 3

Abstract

 ABSTRACT: In this article, a number of posterior probabilistic based equations were introduced to detect the effect of controlling the correlation between variables on the behavior of feed forward neural network weights. In this paper it was proofed that, under certain assumptions, in a feed forward neural network with backprobagation learning algorithm, the correlation between the input variables on one side and the target variable on the other, is directly proportional to the values of the connection weights from the input layer to the output layer through the hidden layer.

Authors and Affiliations

Hazem Migdady, , Yousef Talafha, , Hussam Alrabaiah

Keywords

Related Articles

 Survey on Various Security Issues and prevention method in Vehicular Ad hoc Networks

 Vehicular adhoc networks are a subset of mobile adhoc networks. This forms wireless network between vehicles for better communications and to break away from the hindrance caused by attackers. Various security ar...

Virtual image rendering with stationary RGB color correction for mirror image

Digital image processing which concurrently process the stationary images, and RGB colour corrections in our proposed framework.In this paper we introduce two kind of algorithms.The first one is contour algorithm.Th...

 Breast Cancer Monitoring and Tracking System using Soft Computing and Expert System for Healthcare Support

 Abstract:The decision process for selecting the best-suited follow-up treatment for suspected breast cancer cases are strongly dependent upon the correct diagnosis and assessment of the breast cancer risk. This stu...

Driving Simulator

  Abstract: Simulators which are developed by suppliers individually for specific application purpose are known as Driving Simulator. Most of the conventional driving simulators provide some flexibility for construc...

Home Automation Using Mobile Communication

Abstract: In the present paper we have designed and implemented a typical and simple Home Automation System using GSM (Global System for Mobile Communication) technology that controls electrical devices at home from a re...

Download PDF file
  • EP ID EP147416
  • DOI 10.9790/0661-16323641
  • Views 112
  • Downloads 0

How To Cite

Hazem Migdady, , Yousef Talafha, , Hussam Alrabaiah (2014).  Controlling the Behavior of a Neural Network Weights Using Variables Correlation and Posterior Probabilities Estimation. IOSR Journals (IOSR Journal of Computer Engineering), 16(3), 36-41. https://europub.co.uk./articles/-A-147416