Markov Constraints for Generating Texts with Style

This thesis addresses the issue of generating texts in the style of an existing author, that also satisfy structural constraints imposed by the genre of the text. Although Markov processes are known to be suitable for representing style, they are difficult to control in order to satisfy non-local properties, such as structural constraints, that require long distance modeling. The framework of Constrained Markov Processes allows to precisely generate texts that are consistent with a corpus, while being controllable in terms of rhymes and meter. Controlled Markov processes consist in reformulating Markov processes in the context of constraint satisfaction. The thesis describes how to represent stylistic and structural properties in terms of constraints in this framework and how this approach can be used for the generation of lyrics in the style of 60 differents authors An evaluation of the desctibed method is provided by comparing it to both pure Markov and pure constraint-based approaches. Finally the thesis describes the implementation of an augmented text editor, called Perec. Perec is intended to improve creativity, by helping the user to write lyrics and poetry, exploiting the techniques presented so far.

Italian texts as networks: Topological measurements, Zipf and Heaps' laws

In questa tesi si è studiato un corpus di importanti testi della letteratura Italiana utilizzando la teoria dei network. Le misure topologiche tipiche dei network sono state calcolate sui testi letterari, poi sono state studiate le loro distribuzioni e i loro valori medi, per capire quali di esse possono distinguere un testo reale da sue modificazioni. Inoltre si è osservato come tutti i testi presentino due importanti leggi statistiche: la legge di Zipf e quella di Heaps.

A suggested translation of art related texts

The thesis is the translation of art related texts, translated for a sculptor of Bologna. The comment to the translation is based on functional theories.

Entity extraction e text classification come strumenti per la sentiment analysis di testi giornalistici

In this thesis we are going to talk about technologies which allow us to approach sentiment analysis on newspapers articles. The final goal of this work is to help social scholars to do content analysis on big corpora of texts in a faster way thanks to the support of automatic text classification.

Prediction accuracy of young and middle-aged adults in memory for familiar and unfamiliar texts

This study investigated the influence of age, familiarity, and level of exposure on the metamemorial skill of prediction accuracy on a future test. Young (17 to 23 years old) and middle-aged adults (35 to 50 years old) were asked to predict their memory for text material. Participants made predictions on a familiar text and an unfamiliar text, at three different levels of exposure to each. The middle-aged adults were superior to the younger adults at predicting performance. This finding indicates that metamemory may increase from youth to middle age. Other findings include superior prediction accuracy for unfamiliar compared to familiar material, a result conflicting with previous findings, and an interaction between level of exposure and familiarity that appears to modify the main effects of those variables.