The aim of this lab is the construction of a Machine Learning algorithm to analyse a set of around 3000 technical papers and classify them with respect to their topic. The pre-processing of the data is executed via a Python code on Azure Notebooks to vectorize the text via word tokenisation and embedding. The pre-processed text is registered as an Azure Dataset and is then fed to Azure Automated ML. This feature will find the most suitable ML algorithm according to the accuracy of the model.