Comparison of machine learning algorithms on data from the nuclear industry

With the increase in computing power and the rise of “big data”, many machine learning algorithms have been developed or given a new lease of life. Their application is proved to be really efficient in various fields. However their implementation in the industrial sector, such as nuclear power generation, seems less widespread. This paper presents a comparison of several techniques on a case study from the nuclear industry. Their performance on the real dataset are compared and a discussion is proposed on their practical use, advantages and disadvantages, precaution for use and relevance in an industrial context.