Companies and Artificial Intelligence: An Example of Clustering with Orange

Data mining is a complex process that uses machine learning algorithms to extract “information” from large masses of data (big data). In turn, Machine Learning is a research area of Artificial Intelligence aimed at creating systems (algorithms) able to automatically learn from the experience (training). Learning can be: (1) supervised: the system is provided with input and output data so that it can learn the link between them. The purpose is to identify a general rule that connects the input data with the output data, so that you can reuse that rule. (2) unsupervised: only data are provided to the system without any indication of the desired output. The aim is to go back to hidden patterns and models, identifying a logical structure in the inputs without making it explicit in advance. The supervised and not supervised methods are related to the type of learning and among these one of the most used is the cluster analysis that is a collection of instances (units or records) such that instances of the same cluster are similar to each other and instances of different clusters are dissimilar. After a quick descriptive analysis, this study aims to identify groups of Italian companies that use Artificial Intelligence systems distinguished by sectors of economic activity, purposes and business areas of adoption exploiting the potential of Orange which is a free software for data mining that creates particularly attractive and interesting data visualization without the need for a lot of prior knowledge.