TY - CONF T1 - The Epistle to Cangrande Through the Lens of Computational Authorship Verification T2 - New Trends in Image Analysis and Processing – ICIAP 2019 Y1 - 2019 A1 - Silvia Corbara A1 - Moreo, Alejandro A1 - Sebastiani, Fabrizio A1 - Tavoni, Mirko ED - Cristani, Marco ED - Prati, Andrea ED - Lanz, Oswald ED - Messelodi, Stefano ED - Sebe, Nicu AB - The Epistle to Cangrande is one of the most controversial among the works of Italian poet Dante Alighieri. For more than a hundred years now, scholars have been debating over its real paternity, i.e., whether it should be considered a true work by Dante or a forgery by an unnamed author. In this work we address this philological problem through the methodologies of (supervised) Computational Authorship Verification, by training a classifier that predicts whether a given work is by Dante Alighieri or not. We discuss the system we have set up for this endeavour, the training set we have assembled, the experimental results we have obtained, and some issues that this work leaves open. JF - New Trends in Image Analysis and Processing – ICIAP 2019 PB - Springer International Publishing CY - Cham SN - 978-3-030-30754-7 ER - TY - CONF T1 - Explaining multi-label black-box classifiers for health applications T2 - International Workshop on Health Intelligence Y1 - 2019 A1 - Cecilia Panigutti A1 - Guidotti, Riccardo A1 - Monreale, Anna A1 - Pedreschi, Dino KW - Explainable Machine Learning KW - Healthcare AB - Today the state-of-the-art performance in classification is achieved by the so-called “black boxes”, i.e. decision-making systems whose internal logic is obscure. Such models could revolutionize the health-care system, however their deployment in real-world diagnosis decision support systems is subject to several risks and limitations due to the lack of transparency. The typical classification problem in health-care requires a multi-label approach since the possible labels are not mutually exclusive, e.g. diagnoses. We propose MARLENA, a model-agnostic method which explains multi-label black box decisions. MARLENA explains an individual decision in three steps. First, it generates a synthetic neighborhood around the instance to be explained using a strategy suitable for multi-label decisions. It then learns a decision tree on such neighborhood and finally derives from it a decision rule that explains the black box decision. Our experiments show that MARLENA performs well in terms of mimicking the black box behavior while gaining at the same time a notable amount of interpretability through compact decision rules, i.e. rules with limited length. JF - International Workshop on Health Intelligence PB - Springer UR - https://link.springer.com/chapter/10.1007/978-3-030-24409-5_9 ER - TY - JOUR T1 - ECHO CHAMBERS E POLARIZZAZIONE Uno sguardo critico sulla diffusione dell’informazione nei social network JF - The Lab’s Quarterly Y1 - 2018 A1 - Costantino Carugno A1 - Tommaso Radicioni AB - Understanding the algorithms that contribute to the formation of our daily reality requires an in-depth look at how information is disseminated in online social networks (OSN). In this article, we will observe how news propagation is restricted by the presence of virtual borders that limit the interaction between users. This phenomenon, known as "echo chamber" formation, has the effect of polarizing the public debate on conflicting positions. Inside an echo chamber, information is not conveyed through a horizontal exchange between users, but due to the presence of like or follower aggregators, called hubs. This analysis will be carried out considering a casestudy in two of the main OSNs: Facebook and Twitter. From the study of user interaction networks we will observe how the algorithmic choices made are crucial to the polarization of the debate around a topic of discussion. VL - 20 UR - https://thelabsquarterly.files.wordpress.com/2019/04/2018.4-the-labs-quarterly-7.-costantino-carugno-tommaso-radicioni-1.pdf IS - 4 ER -