Load forecasting in Smart Grids (SG) is a major module of current energy management systems, that play a vital role in optimizing resource allocation, improving grid stability, and assisting the combination of renewable energy sources (RES). It contains the predictive of electricity consumption forms over certain time intervals. Load Forecasting remains a stimulating task as load data has exhibited changing patterns because of factors such as weather change and shifts in energy usage behaviour. The beginning of advanced data analytics and machine learning (ML) approaches; particularly deep learning (DL) has mostly enhanced load forecasting accuracy. Deep neural networks (DNNs) namely Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) have achieved popularity for their capability to capture difficult temporal dependencies in load data. This study designs a Short-Load Forecasting scheme using a Hybrid Deep Learning and Beluga Whale Optimization (LFS-HDLBWO) approach. The major intention of the LFS-HDLBWO technique is to predict the load in the SG environment. To accomplish this, the LFS-HDLBWO technique initially uses a Z-score normalization approach for scaling the input dataset. Besides, the LFS-HDLBWO technique makes use of convolutional bidirectional long short-term memory with an autoencoder (CBLSTM-AE) model for load prediction purposes. Finally, the BWO algorithm could be used for optimal hyperparameter selection of the CBLSTM-AE algorithm, which helps to enhance the overall prediction results. A wide-ranging experimental analysis was made to illustrate the better predictive results of the LFS-HDLBWO method. The obtained value demonstrates the outstanding performance of the LFS-HDLBWO system over other existing DL algorithms with a minimum average error rate of 3.43 and 2.26 under FE and Dayton grid datasets, respectively.

The recent proliferation of ubiquitous computing technologies has led to the emergence of urban computing that aims to provide intelligent services to inhabitants of smart cities. Urban computing deals with enormous amounts of data collected from sensors and other sources in a smart city. In this article, we investigated and highlighted the role of urban computing in sustainable smart cities. In addition, a taxonomy was conceived that categorized the existing studies based on urban data, approaches, applications, enabling technologies, and implications. In this context, recent developments were elucidated. To cope with the engendered challenges of smart cities, we outlined some crucial use cases of urban computing. Furthermore, prominent use cases of urban computing in sustainable smart cities (e.g., planning in smart cities, the environment in smart cities, energy consumption in smart cities, transportation in smart cities, government policy in smart cities, and business processes in smart cities) for smart urbanization were also elaborated. Finally, several research challenges (such as cognitive cybersecurity, air quality, the data sparsity problem, data movement, 5G technologies, scaling via the analysis and harvesting of energy, and knowledge versus privacy) and their possible solutions in a new perspective were discussed explicitly.

Credit risk plays a major role in the banking industry business. Banks' main activities involve granting loan, credit card, investment, mortgage, and others. Credit card has been one of the most booming financial services by banks over the past years. However, with the growing number of credit card users, banks have been facing an escalating credit card default rate. As such data analytics can provide solutions to tackle the current phenomenon and management credit risks. This paper provides a performance evaluation of credit card default prediction. Thus, logistic regression, rpart decision tree, and random forest are used to test the variable in predicting credit default and random forest proved to have the higher accuracy and area under the curve. This result shows that random forest best describe which factors should be considered with an accuracy of 82 % and an Area under Curve of 77 % when assessing the credit risk of credit card customers.

The detection and recognition and then conversion of the characters in an image into a text are called optical character recognition (OCR). A distinctive type of OCR is used to process Arabic characters, namely, Arabic OCR. OCR is increasingly used in many applications where this process is preferred to automatically perform a process without human association. The Quranic text contains two elements, namely, diacritics and characters. However, processing these elements may cause malfunction to the OCR system and reduce its level of accuracy. In this paper, a new method is proposed to check the similarity and originality of Quranic content. This method is based on a combination of Quranic diacritic and character recognition techniques. Diacritic detections are performed using a region-based algorithm. An optimization technique is applied to increase the recognition ratio. Moreover, character recognition is performed based on the projection method. An optimization technique is applied to increase the recognition ratio. The result of the proposed method is compared with the standard Mushaf al Madinah benchmark to find similarities that match with texts of the Holy Quran. The obtained accuracy was superior to the other tested K-nearest neighbor (knn) algorithm and published results in the literature. The accuracies were 96.4286% and 92.3077% better in the improved knn algorithm for diacritics and characters, respectively, than in the knn algorithm