October 12th Aula Fermi
speaker Giambattista Albora
Relatedness is a measure of the affinity between an activity and a location. In the Economic Complexity framework it is used to quantify how much a country is close to the export of a product. Since knowing which products are within the reach of a country is a powerful tool to increase its diversification and with it its fitness, relatedness is a key tool for institutions and policy makers, and a driver for investments. In my presentation I will take in consideration different relatedness measures that have been defined in the last decade. The traditional way to measure the relatedness is based on a one mode projection of the bipartite country-product network on the layer of the products and it has been introduced by Hidalgo et al with the Product Space in 2012 . Our group recently introduced a new approach based on machine learning  which has been shown to provide a better assessment of the relatedness. In addition to presenting and comparing all the different relatedness measures showing that the machine learning approach outperforms the network one, I will show that also a simple transposition of the problem brings to a better assessment of the relatedness. In other words, when using a network approach, if one looks to the similarity among countries instead of the one among products the results are improved, and with the same reasoning also the results of the machine learning approach can be improved. Finally I will show how networks can be used to further improve the results of a machine learning algorithm, in particular when the considered network is the Sapling Similarity, a new network model on which we are recently working that allows the possibility to have negative links.
 Hidalgo CA, Klinger B, Barab´asi AL, Hausmann R. The product space conditions the development of nations. Science. 2007;317(5837):482–487.
 Albora G, Pietronero L, Tacchella A, Zaccaria A. Product Progression: a machine learning approach to forecasting industrial upgrading. arXiv preprint arXiv:210515018. 2021