Machine Learning for Data Flow Processing in Learning Process

Abstract

Artificial intelligence is the set of information processing technologies aimed at conferring cognitive abilities comparable to that of humans to computers, machines or robots, through mathematical functions, algorithms but also modes of representation of the real environment. There are generally two main approaches in AI: symbolist approaches, where knowledge and reasoning are represented by a mathematical formulation or logic (with symbols) and connectionist approaches, where cognitive function is achieved by assembling formal, mathematical neurons, connected to each other in more or less complex networks (neural networks). Faced with the difficulties encountered by symbolist approaches to formalize robust reasoning in a real world with many exceptions, and also thanks to the opportunity represented by accumulated data (big data), learning cognitive functions from data becomes a promising option. This is the whole point of machine learning. In this research work, we are mainly interested in the methods of machine learning. A practical case will be discussed in this paper.