Towards big industrial data mining through explainable automated machine learning

Industrial systems resources are capable of producing large amount data. These data often in heterogeneous formats and distributed, yet they provide means to mine the information which can allow deployment intelligent management tools for production activities. For this purpose, it is necessary be able implement knowledge extraction prediction processes using Artificial Intelligence (AI) models, but selection configuration intended AI models tend increasingly complex a non-expert user. In paper, we present an approach software platform that may industrial actors, who usually not familiar with AI, select configure algorithms optimally adapted their needs. Hence, essentially based on automated machine learning. The resulting effectively enables better choice among combination hyper-parameters configurations. It also makes possible features explainability thus increasing acceptability these practicing community users. proposed has been applied field predictive maintenance. Current tests analysis more than 360 databases from subjected field.