A discussion took place in the Faculty of Computer Science and Information Technology, under the supervision of the Dean of the College, Dr. Salma Abdel-Baqi Mahmoud, the respected master’s thesis entitled “Forecasting crude oil prices in Iraq using neural networks.”
For the student Suhair Aziz Hafez, under the supervision of Professor Maysa Abdel Karim, in the Department of Computer Information Systems, she discussed it
Building a cascade network model for forecasting Iraqi crude oil prices
The model consists of two neural networks, operating in series. The first network is the front-end network (GBNet) that works to find the most influential factor on the price of oil in Iraq among a group of factors that include (the cost of producing a barrel of oil, the quantity of production, global demand and supply, the exchange rate of the dollar against the Iraqi dinar and geopolitical events) . The second network is a recurrent neural network (ConvolutionalLSTM) that was used to predict the price of Iraqi oil based on the prices of Iraqi crude oil in previous years, in addition to the value of the outputs of the first network. The committee in charge of discussing the thesis in the presence of the faculty in the postgraduate hall recommended the acceptance of the thesis with some minor modifications, with a grade of very good high
