No abstract

Nowadays, learning-based modeling system is adopted to establish an accurate prediction model for renewable energy resources. Computational Intelligence (CI) methods have become significant tools in production and optimization of renewable energies. The complexity of this type of energy lies in its coverage of large volumes of data and variables which have to be analyzed carefully. The present study discusses different types of Deep Learning (DL) algorithms applied in the field of solar and wind energy resources and evaluates their performance through a novel taxonomy. It also presents a comprehensive state-of-the-art of the literature leading to an assessment and performance evaluation of DL techniques as well as a discussion about major challenges and opportunities for comprehensive research. Based on results, differences on accuracy, robustness, precision values as well as the generalization ability are the most common challenges for the employment of DL techniques. In case of big dataset, the performance of DL techniques is significantly higher than that for other CI techniques. However, using and developing hybrid DL techniques with other optimization techniques in order to improve and optimize the structure of the techniques is preferably emphasized. In all cases, hybrid networks have better performance compared with single networks, because hybrid techniques take the advantages of two or more methods for preparing an accurate prediction. It is recommended to use hybrid methods in DL techniques. INDEX TERMS Big dataset, deep learning, modeling, optimizing, solar energy, wind energy. ACRONYMS USED FREQUENTLY IN THIS WORK GHG Greenhouse gas LSTM Long short-term memory Network FL Fuzzy logic SAE Stacked auto-encoder DL Deep learning DRL Deep reinforcement learning CI Computational intelligent WNN wavelet neural network DBN Deep belief network DRWNN diagonal recurrent wavelet neural network RBM Restricted Boltzmann machine ANFIS Adaptive neuro fuzzy inference system The associate editor coordinating the review of this manuscript and approving it for publication was Ton Do. RBF Radial basis function EC Evolutionary computation CNN Convolutional neural network MLP Multi layered perceptron TDNN Time delay neural network NARNN Nonlinear auto regressive neural network FFNN Feed-forward neural network CPRS Continuous ranking probability score (ANNs Artificial neural networks SVM Support vector machine MRBM Multilayer Restricted Boltzmann Machine BPNN Back Propagation Neural Network WT Wavelet transform QR Quintile regression

No abstract

Prediction of stock groups values has always been attractive and challenging for shareholders. This paper concentrates on the future prediction of stock market groups. Four groups named diversified financials, petroleum, non-metallic minerals and basic metals from Tehran stock exchange are chosen for experimental evaluations. Data are collected for the groups based on ten years of historical records. The values predictions are created for 1, 2, 5, 10, 15, 20 and 30 days in advance. The machine learning algorithms utilized for prediction of future values of stock market groups. We employed Decision Tree, Bagging, Random Forest, Adaptive Boosting (Adaboost), Gradient Boosting and eXtreme Gradient Boosting (XGBoost), and Artificial neural network (ANN), Recurrent Neural Network (RNN) and Long short-term memory (LSTM). Ten technical indicators are selected as the inputs into each of the prediction models. Finally, the result of predictions is presented for each technique based on three metrics. Among all algorithms used in this paper, LSTM shows more accurate results with the highest model fitting ability. Also, for tree-based models, there is often an intense competition between Adaboost, Gradient Boosting and XGBoost.