[ Bio ]

Riccardo Ricciardi comes from a town near Naples, in the south of Italy, but studied and started working in the north of the country.
In January, he will defend his PhD Thesis in Statistics, titled "Statistical analysis of grouped text documents", at the Department of Economics and Management of the University of Brescia. Since November,  Riccardo Ricciardi have been working as a Research Fellow in the same Department, involved in the MICS – Made in Italy Circular and Sustainable Extended Partnership between Universities, Research Centers, and Enterprises financed by the Ministry of University and Research. His work in that project is dedicated to studying the semantics of Made-in-Italy products, and how consumers perceive the transition of Made-in-Italy towards circularity and sustainability.

Outside Academia,  Riccardo Ricciardi funded Metro-Polis, an analytics consulting start-up company, in the city of Turin, and DataBeers Brescia, the Brescia's branch of the DataBeers format, funded in Madrid, where, in a relaxed environment with free beers, a night of talks about Data Science is offered.

[ Abstract ]

Statistical analysis of grouped text documents

The seminar presents a collection of studies in the context of statistical analysis of textual data. It shows two applications and one methodological study of well-known approaches for text document representations, including Vector Space Models, Paragraph2Vec, and BERT embeddings. The studies share the problem of dealing with grouped documents. Two domains of application are explored: social media and cultural tourism. About the former, a study is proposed about self-presentation among diverse groups of individuals on the StockTwits platform, where stock markets are the dominant topic. About the latter, the thesis focuses on online reviews of cultural attractions in the Italian city of Brescia.

Lastly, the seminar presents a methodological study examining the group-specificity of words, analyzing various group-specificity estimators proposed in the literature on simulated data.