Data Flow Analysis Using Matrix Calculations

In recent decades, many household items have become computerized, collecting data and sending it for further processing. Consequently, the number of data flow structure requirements to analyze data in real-time is steadily increasing. Besides, sensors, remote sensing, social networks, and other fields of activity generate data in the form of arrays. However, the projects use existing statistical data streams for matrix calculations. From a theoretical point of view, matrices and matrix transformations are widely used to solve many problems in computer science, such as machine learning, electrical circuit analysis, and image and graph processing. Thus, it will be possible to build different algorithms for many spheres, based on the basic matrix operations. Thus, the main goals of this work are to provide matrix calculations for data flow analysis. To do this, you need to add matrix data structures for built-in support for matrices as stream elements and matrix operations. Besides, these matrix classes must be used for operators that do not alter or storing state.