New markov chain based methods for single and cross-domain sentiment classification

Nowadays communication is switching from a centralized scenario, where communication media like newspapers, radio, TV programs produce information and people are just consumers, to a completely different decentralized scenario, where everyone is potentially an information producer through the use of social networks, blogs, forums that allow a real-time worldwide information exchange. These new instruments, as a result of their widespread diffusion, have started playing an important socio-economic role. They are the most used communication media and, as a consequence, they constitute the main source of information enterprises, political parties and other organizations can rely on. Analyzing data stored in servers all over the world is feasible by means of Text Mining techniques like Sentiment Analysis, which aims to extract opinions from huge amount of unstructured texts. This could lead to determine, for instance, the user satisfaction degree about products, services, politicians and so on. In this context, this dissertation presents new Document Sentiment Classification methods based on the mathematical theory of Markov Chains. All these approaches bank on a Markov Chain based model, which is language independent and whose killing features are simplicity and generality, which make it interesting with respect to previous sophisticated techniques. Every discussed technique has been tested in both Single-Domain and Cross-Domain Sentiment Classification areas, comparing performance with those of other two previous works. The performed analysis shows that some of the examined algorithms produce results comparable with the best methods in literature, with reference to both single-domain and cross-domain tasks, in $2$-classes (i.e. positive and negative) Document Sentiment Classification. However, there is still room for improvement, because this work also shows the way to walk in order to enhance performance, that is, a good novel feature selection process would be enough to outperform the state of the art. Furthermore, since some of the proposed approaches show promising results in $2$-classes Single-Domain Sentiment Classification, another future work will regard validating these results also in tasks with more than $2$ classes.

Sistemi di Regolazione dell'Alimentazione Affidabili per Processori Multi-Core

I moderni processori multi-core ad elevate prestazioni sono alimentati da regolatori di tensione integrati direttamente sul chip. Questi regolatori forniscono a ciascun power domain la tensione ottimale sulla base della sua attività, monitorata da una Power Control Unit. Questo consente da un lato di ottenere una riduzione dei consumi, dall'altro di avere un boost delle prestazioni in particolari contesti. Tali regolatori integrati sul die sono affetti da guasti e fenomeni di aging, che possono compromettere il corretto funzionamento del circuito. Questi problemi non sono tollerabili in contesti caratterizzati da esigenze di elevata reliability, come l'autonomous driving. Dunque, è stato sviluppato un monitor per rivelare on-line eventuali guasti che possono verificarsi durante il normale funzionamento sul campo. In caso di guasto il monitor è in grado di dare un'indicazione d'errore, che può essere utilizzata per attivare delle procedure di recovery. La soluzione proposta, basata su un approccio completamente differente rispetto a quello suggerito dallo standard ISO 26262, beneficia, rispetto a quest'ultima, di costi nettamente inferiori e prestazioni superiori. Il monitor può essere calibrato automaticamente per compensare le variazioni dei parametri di processo ed i fenomeni di aging che possono affliggere il monitor stesso. È stata verificata la self-checking ability del monitor rispetto a guasti di tipo transistor stuck-on, transistor stuck-open e bridging resistivo, risultando Totally Self-Checking rispetto all'insieme di guasti considerato.

Hierarchical Multi-Label Text Classification in a Low-Resource Setting

In this thesis we address a multi-label hierarchical text classification problem in a low-resource setting and explore different approaches to identify the best one for our case. The goal is to train a model that classifies English school exercises according to a hierarchical taxonomy with few labeled data. The experiments made in this work employ different machine learning models and text representation techniques: CatBoost with tf-idf features, classifiers based on pre-trained models (mBERT, LASER), and SetFit, a framework for few-shot text classification. SetFit proved to be the most promising approach, achieving better performance when during training only a few labeled examples per class are available. However, this thesis does not consider all the hierarchical taxonomy, but only the first two levels: to address classification with the classes at the third level further experiments should be carried out, exploring methods for zero-shot text classification, data augmentation, and strategies to exploit the hierarchical structure of the taxonomy during training.