Showing papers by "Adnan Khashman published in 2020"
27 Aug 2020
TL;DR: In this paper, an intelligent deep learning model that is designed and trained to anticipate people's expectations regarding reaching a political settlement for the ongoing Cyprus conflict is discussed. And the proposed method based on deep learning architecture, can be successfully applied to predict similar real-life applications where people's reactions can be recorded, digitized and used to predict their perceptions.
Abstract: A novel application of artificial intelligence in social sciences is discussed in this paper. The work describes an intelligent deep learning model that is designed and trained to anticipate people’s expectations regarding reaching a political settlement for the ongoing Cyprus conflict. Our model utilizes deep learning neural networks architecture to associate people’s information and views on important issues related to the ongoing peace process mediations, together with their expectation for signing a peace deal by the end of year 2020. We collected the database through completed surveys and polls by a group of university students studying international relations in Northern Cyprus. The obtained database undergoes digital coding prior to using it to train and test the proposed prediction model. A deep learning fully connected dense neural network model (DNN) with five hidden layers is designed, implemented, and its performance is then compared to that of the classic back propagation neural network. The obtained results suggest that the proposed novel method based on deep learning architecture, can be successfully applied to predict similar real-life applications where people’s reactions can be recorded, digitized and used to predict their perceptions.