Among the many music information retrieval (MIR) tasks, music genre classification is noteworthy. The categorization of music into different groups that came to existence through a complex interplay of cultures, musicians, and various market forces to characterize similarities between compositions and organize collections is known as a music genre. The past researchers extracted various hand-crafted features and developed classifiers based on them. But the major drawback of this approach was the requirement of field expertise. However, in recent times researchers, because of the remarkable classification accuracy of deep learning models, have used similar models for MIR tasks. Convolutional Neural Net- work (CNN), Recurrent Neural Network (RNN), and the hybrid model, Convolutional - Recurrent Neural Network (CRNN), are such prominently used deep learning models for music genre classification along with other MIR tasks and various architectures of these models have achieved state-of-the-art results. In this study, we review and discuss three such architectures of deep learning models, already used for music genre classification of music tracks of length of 29-30 seconds. In particular, we analyze improved CNN, RNN, and CRNN architectures named Bottom-up Broadcast Neural Network (BBNN) [1], Independent Recurrent Neural Network (IndRNN) [2] and CRNN in Time and Frequency dimensions (CRNN- TF) [3] respectively, almost all of the architectures achieved the highest classification accuracy among the variants of their base deep learning model. Hence, this study holds a comparative analysis of the three most impressive architectural variants of the main deep learning models that are prominently used to classify music genre and presents the three architecture, hence the models (CNN, RNN, and CRNN) in one study. We also propose two ways that can improve the performances of the RNN (IndRNN) and CRNN (CRNN-TF) architectures.

This document proposes a model of the process of atmospheric discharge of overhead lines followed by an electric arc. The intensity of atmospheric discharges, followed by electric arc and destruction, is determined by the difference in potential and current. Such a structure and form of discharge make it difficult to analyze the transient process and obtain adequate solutions. That is why the model of the transient process in the electric arc of the switch under the conditions of interruption of the AC circuit is specially analyzed. Simple equivalent schemes for the analysis of phenomena with given values of linear parameters are presented, which are very simple to apply. All influential parameters by which the overvoltage values in the model can be estimated were also taken into account. The evaluation of the proposed model was performed using the adapted MATLAB program psbsurlightcuurent for atmospheric electrical discharge, which contains a high frequency current source. The verified simulation method was used to verify the results as part of a method derived from artificial intelligence algorithms. The process simulation program, the obtained voltage and current diagrams confirm the application of the simulation algorithm model.

In recent years, with the advancement of the internet, social media is a promising platform to explore what going on around the world, sharing opinions and personal development. Now, Sentiment analysis, also known as text mining is widely used in the data science sector. It is an analysis of textual data that describes subjective information available in the source and allows an organization to identify the thoughts and feelings of their brand or goods or services while monitoring conversations and reviews online. Sentiment analysis of Twitter data is a very popular research work nowadays. Twitter is that kind of social media where many users express their opinion and feelings through small tweets and different machine learning classifier algorithms can be used to analyze those tweets. In this paper, some selected machine learning classifier algorithms were applied on crawled Twitter data after applying different types of preprocessors and encoding techniques, which ended up with satisfying accuracy. Later a comparison between the achieved accuracies was showed. Experimental evaluations show that the Neural Network Classifier' algorithm provides a remarkable accuracy of 81.33% compared with other classifiers.

In this day and age, the rise in technological advancements has the potential to improve and transform our lives every day. The rapid technology innovation can have a great impact on our business operations. Currently, Cloud computing services are popular and offer a wide range of opportunities for their customers. This paper presents a survey on a more recent computing architecture paradigm known as Fog Computing. Fog networking is a beneficial solution that offers the greater facility of data storage, enhanced computing, and networking resources. This new concept of fog complements cloud solution by facilitating its customers with better security, real-time analysis improved efficiency. To get a clear picture and understanding of how fog computing functions, we have performed an extensive literature review. We also presented a comparative study of fog computing with cloud and grid computing architectures. In this study, we have conducted a survey that led us to the conclusion that fog computing solution is still not applicable and implemented in most of the IoT industries due to the lack of awareness and the high architecture’s cost. Results of the study also indicate that optimized data storage and security are a few of the factors that can motivate organizations to implement the Fog computing architecture. Furthermore, the challenges related to fog computing solution are reviewed for progressive developments in the future.