Scholarlink

In recent years, interest in the topic of forecasting has a plus, the forecasting process has received and continues to receive great and increasing interest from researchers and decision-makers, and the methods used to improve it develop from time to time, forecasting also reduces some of the risks facing oil institutions in particular in the future. In the current period, oil forecasting is considered very important as a result of the situation. Economic and political in Iraq, as the presence of economic crises threatens the continued collapse of the economy in Iraq, modern and advanced methods appeared, namely artificial intelligence techniques. Among these techniques are artificial neural network models that have proven their superiority in this because they are distinguished by the ability to learn and train avoiding assumptions about the nature of the time series in comparison with traditional statistical models. Artificial neural networks have become widely used in various applications because of their ability to form accurate and acceptable predictions in cases where there are complex relationships between inputs and outputs and do not require a model with specific specifications, as neural networks have the property of dealing with any model through the network’s self-learning. The objective of this present study is to elucidate various statistical methodologies employed in predicting future oil production and pricing trends in Iraq up to the year 2035. A neural network was built for the purpose of predicting oil production and prices by determining its inputs and outputs and determining the components of the hidden layer in it, and then training the network using production and prices. Oil for the period extending from (1976 to 2022). Through experimental analysis, it was found that the value of the mean square error (MSE = 0.01) is the lowest error value, in addition to the high value of the coefficient of determination (R square = 0.774) and accuracy (Accuracy = 85%). The results show that the artificial neural network has proven its efficiency in prediction accuracy as it has the lowest value of the mean square errors (MSE). This reflects the ability of the artificial neural network technology to predict oil prices and production.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Using Artificial Neural Networks in Prediction with Application

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Zainab Muslim Nasser Al-Ali,
  • Sahera Hussein Zain Al-Thalabi

摘要

In recent years, interest in the topic of forecasting has a plus, the forecasting process has received and continues to receive great and increasing interest from researchers and decision-makers, and the methods used to improve it develop from time to time, forecasting also reduces some of the risks facing oil institutions in particular in the future. In the current period, oil forecasting is considered very important as a result of the situation. Economic and political in Iraq, as the presence of economic crises threatens the continued collapse of the economy in Iraq, modern and advanced methods appeared, namely artificial intelligence techniques. Among these techniques are artificial neural network models that have proven their superiority in this because they are distinguished by the ability to learn and train avoiding assumptions about the nature of the time series in comparison with traditional statistical models. Artificial neural networks have become widely used in various applications because of their ability to form accurate and acceptable predictions in cases where there are complex relationships between inputs and outputs and do not require a model with specific specifications, as neural networks have the property of dealing with any model through the network’s self-learning. The objective of this present study is to elucidate various statistical methodologies employed in predicting future oil production and pricing trends in Iraq up to the year 2035. A neural network was built for the purpose of predicting oil production and prices by determining its inputs and outputs and determining the components of the hidden layer in it, and then training the network using production and prices. Oil for the period extending from (1976 to 2022). Through experimental analysis, it was found that the value of the mean square error (MSE = 0.01) is the lowest error value, in addition to the high value of the coefficient of determination (R square = 0.774) and accuracy (Accuracy = 85%). The results show that the artificial neural network has proven its efficiency in prediction accuracy as it has the lowest value of the mean square errors (MSE). This reflects the ability of the artificial neural network technology to predict oil prices and production.