My research is focused on the automation of annotation process for training and test set building. In the last decades, the interest on machine learning has greatly increased, because it is able to perform various tasks and it is useful in many applications, such as automatic speech recognition, text categorization, computer-aided diagnosis, tracking and document classification. These applications are basic pillars in Industry 4.0 paradigm. Supervised machine learning process starts from the annotation task of each element of a dataset to build the training and test set, but this task is performed manually. To perform a machine learning process a huge dataset is necessary to ensure a high-quality result of the learning process. The need of automatic annotation arises in current landscape to enhance the performances of machine learning algorithm and to invest the human intelligence in other more valuable tasks. I will create a series of software applications to support the phase of annotation, with the attempt to automate the labelling tasks that have important consequences on both academy and industry. In my first year of PhD course, I want to study the machine learning processes and the annotation process for training and test set building with a focus in different industrial applications. In second year of my PhD, I want to measure the degree of formalization in labelling tasks in different industrial fieldsand formalize the annotation task rules. Finally, in third year, I want to design a model for automatic labelling elements and develop a set of tools for applying the model to different industrial applications.