Loading...

Design and Efficient Implementation of Neural Networks for Solving Graph-based Problems

Mahdipour Araste, Payam | 2018

592 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 51468 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Hashemi, Matin
  7. Abstract:
  8. The extraordinary ability of the human brain to solve various problems has led scientists to simulate models of the human brain. One of these simulated models is artificial neural networks. Today, the power of artificial neural networks is not overlooked. The ability of artificial neural networks to solve various types of issues led us to use the thesis to solve some of the graph-based problems. Quite accurately, this graph-based problem is a matter of identifying the source of rumor in a network. In many graph networks, whether natural networks such as the network of neurons in the human brain or synthetic ones such as the types of social networks, it is possible that a rumor spreads across the network and overwhelms the various components of the network. In this thesis, we intend to find a rumor source among all the network nodes in terms of affecting some of the nodes within the network or not affecting them from the rumor, as well as with the network topology.The ability of artificial neural networks increases with increasing number of neuronal layers.Therefore, we used deep neural networks to solve the sophisticated problem of identifying the rumor source in a network. The proposed deep neural network model identifies about 18% of the rumor source more accurately than the best source detection method. So far,none of the methods of source identification has been used for deep learning. Hence, this thesis can create new branches in source identification methods
  9. Keywords:
  10. Artificial Neural Network ; Deep Learning ; Sources Identification ; Rumor Spreading ; Graph Analysis ; Graph Embeding

 Digital Object List

 Bookmark

No TOC