Augmented Analytics through a Conversational Virtual Advisor: a proof-of-concept in the food industry

Nowadays data are becoming more and more a valuable and sought after resource, which makes it even more crucial to have a working infrastructure for performing analytical processes on the data itself inside of an organization, so the company can extract more information from the data and achieve a better performance. In this whole data analysis context where high quantity of data are constantly produced and updated the need for quicker and more automated analysis becomes necessary. For this purpose, Augmented Analytics represents a new approach for data analysis which aims at augmenting and enhancing the whole analytical process inside of a company by leveraging a series of advanced and innovative technologies, such as Artificial Intelligence, Machine Learning and Natural Language Processing techniques. The main goals of Augmented Analytics are to enhance the overall results obtained by the analysis process, but also to automate most of this process itself, and subsequently transforming it into a more rapid and simplified procedure, making then also the whole data analysis context more broadly accessible. Considering this whole environment and context, this thesis project will focus on the design and implementation of such Augmented Analytics techniques and principles in a real case study of a well-established Italian company with plenty of proprietary data operating in the agri-food industry. These techniques and principles will be implemented through a proposal of a proof-of-concept data analytics platform, called ”Conversational Virtual Advisor”, which aims at providing data analysis results to its users by just needing to get asked questions in natural language, providing then a solution which in overall represents a faster, easier and more automated alternative to the one already present within this specific company context.