Argomenti: continuazione della prima parte,seed choice,buckshot algorithm,soft clustering,expectation maximumization,EM,probabilistically labeled data,Naive Bayes,semi-supervied learning,issue in clustering