Real-Time Big Data Analytics for Live Streaming Video Quality Assessment Using Deep Learning

In recent years, live streaming video has become increasingly popular, necessitating real-time big data analytics to evaluate streaming video content. In this article, it provides a framework for assessing the quality of live broadcast video using big data analytics and deep learning in real time. Our framework comprises various modules, including collection of data, preprocessing, extraction of features, selection of features, and categorization based on deep learning. To determine the quality of the streaming video, we assess a variety of video characteristics, including resolution, bit rate, frame rate, and packet loss rate. Using convolutional neural networks (CNNs), our deep learning models categorise the streaming video quality into various categories. Using real-world streaming video datasets, we evaluate our framework's performance. The results demonstrate that our proposed framework is able to efficiently analyse the quality of streaming video in real-time, outperforming other state-of-the-art frameworks. Our framework provides live streaming platforms with a tool for delivering high-quality streaming video to their consumers.