Analizzare gli Appalti Pubblici usando Machine Learning

Public Procurement is one of the government activities most vulnerable to anomalies. Nowadays a lot of Data Scientists, Researchers and Professors face with the problem of finding suitable solutions regarding these problems. On this Master Thesis, we will focus on using Machine Learning to help us detect anomalies in Public Procurement data. The data-set used throughout this research process, is provided by the Italian National Anti-Corruption Authority (ANAC). We will build a Machine Learning Predictive Model capable of detecting one of these anomalies, in this case the ricorso in the Public Procurement domain. We will also use Natural Language Processing to extract valuable and structured information from the Public Procurement data-set. And finally we will focus on building a Machine Learning Predictive Model, capable of detecting a second type of anomalies, in this case the variazioni in corso d'opera. These Machine Learning Models are just the tip of the iceberg and show us what Machine Learning algorithms are capable of achieving also in such complex domains cases.