High Quality Dataset for Machine Learning in the Business Intelligence Domain

This paper is aimed at showing the relevance and importance of high quality dataset in machine learning within the field of economic intelligence. As open source dataset flourish and algorithm are trained with different and often very narrow data of various kind, in the field of economic intelligence it is important to train machines with proper and high value data to avoid or reduce at maximum false positives, errors and biases of various kind. We propone the case study and the solution offered by Bureau Van Dijk, where economic data are carefully evaluated, organized and its API and REST services could matter in the very near future in the field of data mining and machine learning for economic intelligence purposes.