Tesi di Dottorato
Permanent URI for this communityTesi di Dottorato
Browse
76 results
Search Results
Item Applicazione dei big data nel turismo, marketing ed education(Università della Calabria, 2020-03-18) Giglio, Simona; Critelli, Salvatore; Pantano, PietroIl mondo è attualmente inondato da dati e l’avanzare delle tecnologie digitali amplifica questo fenomeno in modo esponenziale. Tale fenomeno viene etichettato con il concetto di Big Data ovvero le tracce digitali che le nostre attività quotidiane lasciano per effetto dell’uso massiccio dei sistemi ICT (Information Communication Technologies). I Big Data sono diventati il nuovo microscopio che rende “misurabile” la società. Per tali ragioni, la ricerca è incentrata sull’analisi dei Big data, estratti dai social media, da indagini online, da piattaforme di recensioni e da database, attraverso l’applicazione di tecniche e strumenti sviluppati nell’ambito dell’Intelligenza Artificiale. Algoritmi di machine learning, analisi semantica ed analisi statistica sono stati utilizzati per estrarre, dai Big Data, informazione sotto forma di “conoscenza” e “valore”, dimostrando come dati di grandi dimensioni possano fungere da ricca fonte di informazione, da un lato, per comprendere il comportamento dell’utente, parte integrante di una società complessa (conoscenza), e dall’altro, per sostenere i processi decisionali e i servizi forniti agli utenti/consumatori (valore). Il lavoro si caratterizza per un approccio multidisciplinare tra settori differenti quali le scienze sociali, le scienze statistiche e l’informatica. Questo ha permesso di fondare la ricerca sui Big Data nella teoria, e fornire un efficace recupero e analisi dei dati nella pratica. Le tecniche di machine learning sono state applicate per (i) il riconoscimento delle immagini, (ii) per la creazione di cluster, (iii) per l’analisi del testo (sentiment analysis) e (iv) per la profilazione di classi di utenti. Per il riconoscimento delle immagini l’approccio ha richiamato le reti neurali artificiali (deep artificial neural networks), algoritmi e sistemi computazionali ispirati al cervello umano utilizzando le potenzialità del programma Wolfram Mathematica e la disponibilità di dati estratti da social network quali Flickr, Twitter, Instagram ed altre piattaforme come TripAdvisor. Gli strumenti utilizzati nella ricerca hanno permesso di indagare e di rilevare in modo oggettivo dall’analisi di immagini e di testi condivisi sul web, alcuni comportamenti cognitivi degli utenti/consumatori alla base delle loro scelte nonché l’attrattività di una destinazione turistica e la qualità dell’esperienza dell’utente. Lo studio del significato delle parole nel testo ha aperto la strada al web semantico che permette ad un utente di acquisire informazioni approfondite durante una ricerca attraverso un sistema formato da una rete di relazioni e connessioni tra documenti. Partendo dalle ricerche di Ogden e Richards sullo studio del significato e di Jakobson che studiò i processi comunicativi, si è cercato di strutturare e sistematizzare un processo che riflette un atto comunicativo ed informativo tale che un simbolo (immagine) attraverso l’applicazione di un significante (machine learning che si sostituisce al processo mentale proprio dell’uomo) permettesse l’esplicitazione di un referente (oggetto\etichetta) che opportunatamente porta alla trasmissione di un messaggio sotto forma di conoscenza. Il tutto coordinato da un sistema in grado di coniugare fattori differenti in un’ottica interdisciplinare dove l’analisi dei dati combacia perfettamente con la linguistica. Attingendo da studi precedenti, i risultati raggiunti dimostrano che gli algoritmi di analisi dei Big Data quali l’apprendimento automatico contribuiscono da un lato alla comprensione sull’esperienza dell’utente verso un luogo, una destinazione; d’altra parte, la loro analisi fornisce una conoscenza sistematica delle valutazioni dei consumatori su un determinato prodotto o servizio e verso lo sviluppo di una sorta di “intelligenza sociale”. Inoltre i risultati della ricerca propongono come, un approccio più sofisticato al monitoraggio dei social media nel contesto turistico e nel marketing, nonché nel settore dell’education, possa contribuire a migliorare le decisioni strategiche e le politiche operative degli stakeholder nonché ad avere una visione psicologica sugli atteggiamenti e sul comportamento di un ampio spettro di utenti.Item <> costruzione di possibili profili di sostenibilità applicati a scala di quartiere.(2013-11-29) Manfredi, Emilia; Pantano, Pietro; Rossi, FrancescoItem Nano materials and innovative laser-based accelerators for cultural heritage(2017-07-12) Veltri, Simona; Pantano, Pietro; Bonanno, Assunta; Antici, PatrizioUniversità della Calabria, Dipartimento di FisicaItem Nano materials and innovative laser-based accelerators for cultural heritage(2017-07-12) Veltri, Simona; Pantano, Pietro; Bonanno, Assunta; Antici, PatrizioItem Charged-particle distributions and material measurements in ps = 13 TeV pp collisions with the ATLAS Inner Detector(2017-07-14) Cairo, Valentina Maria Martina; Pantano, Pietro; Dell'Acqua, Andrea; Schioppa, MarcoThe Run 2 of the Large Hadron Collider, which began in Spring 2015, offers new challenges to the Experiments with its unprecedented energy scale and high luminosity regime. To cope with the new experimental conditions, the ATLAS Experiment was upgraded during the first long shutdown of the collider, in the period 2013-2014. The most relevant change which occurred in the ATLAS Inner Detector was the installation of a fourth pixel layer, the Insertable B-Layer, at a radius of 33 mm together with a new thinner beam pipe. The Pixel Services, located between the Pixel and SCT detectors, were also modified. Owing to the radically modified ID layout, many aspects of the track reconstruction programs had to be re-optimized. In this thesis, the improvements to the tracking algorithms and the studies of the material distribution in the Inner Detector are described in detail, together with the improvements introduced in the geometry model description in simulation as well as the re-evaluation and the reduction of the systematic uncertainty on the estimate of the track reconstruction efficiency. The results of these studies were applied to the measurement of Charged-Particle Multiplicity in proton–proton collisions at a centre-of-mass energy of 13 TeV. The chargedparticle multiplicity, its dependence on transverse momentum and pseudorapidity and the dependence of the mean transverse momentum on the charged-particle multiplicity are presented for various fiducial phase spaces. The measurements are corrected for detector effects, presented as particle-level distributions and are compared to the predictions of different Monte Carlo event generators. New sets of recommended performance figures along with the related systematic uncertainties were also derived for several aspects of the ATLAS tracking, such as track reconstruction efficiency, fake rate and impact parameter resolution. These recommendations provide information on appropriate working points, i.e. track selection criteria with wellunderstood performance. They apply to physics analyses using Inner Detector tracks in Run 2 data and are important inputs for other objects based on tracks, such as jets. A simulation-based method which uses the tracking recommendations to calibrate light-jets mis-tagged as b-jets it is also presented in the context of the measurement of the crosssection of the W-boson produced in association with b-jets at 13 TeV, together with an overview of the inclusiveW-boson cross-section analysis.Item Charged-particle distributions and material measurements in ps = 13 TeV pp collisions with the ATLAS Inner Detector(2017-07-14) Cairo, Valentina Maria Martina; Pantano, Pietro; Dell'Acqua, Andrea; Schioppa, MarcoThe Run 2 of the Large Hadron Collider, which began in Spring 2015, offers new challenges to the Experiments with its unprecedented energy scale and high luminosity regime. To cope with the new experimental conditions, the ATLAS Experiment was upgraded during the first long shutdown of the collider, in the period 2013-2014. The most relevant change which occurred in the ATLAS Inner Detector was the installation of a fourth pixel layer, the Insertable B-Layer, at a radius of 33 mm together with a new thinner beam pipe. The Pixel Services, located between the Pixel and SCT detectors, were also modified. Owing to the radically modified ID layout, many aspects of the track reconstruction programs had to be re-optimized. In this thesis, the improvements to the tracking algorithms and the studies of the material distribution in the Inner Detector are described in detail, together with the improvements introduced in the geometry model description in simulation as well as the re-evaluation and the reduction of the systematic uncertainty on the estimate of the track reconstruction efficiency. The results of these studies were applied to the measurement of Charged-Particle Multiplicity in proton–proton collisions at a centre-of-mass energy of 13 TeV. The chargedparticle multiplicity, its dependence on transverse momentum and pseudorapidity and the dependence of the mean transverse momentum on the charged-particle multiplicity are presented for various fiducial phase spaces. The measurements are corrected for detector effects, presented as particle-level distributions and are compared to the predictions of different Monte Carlo event generators. New sets of recommended performance figures along with the related systematic uncertainties were also derived for several aspects of the ATLAS tracking, such as track reconstruction efficiency, fake rate and impact parameter resolution. These recommendations provide information on appropriate working points, i.e. track selection criteria with wellunderstood performance. They apply to physics analyses using Inner Detector tracks in Run 2 data and are important inputs for other objects based on tracks, such as jets. A simulation-based method which uses the tracking recommendations to calibrate light-jets mis-tagged as b-jets it is also presented in the context of the measurement of the crosssection of the W-boson produced in association with b-jets at 13 TeV, together with an overview of the inclusiveW-boson cross-section analysis.Item Charged-particle distributions and material measurements in ps = 13 TeV pp collisions with the ATLAS Inner Detector(2017-07-14) Cairo, Valentina Maria Martina; Pantano, Pietro; Schioppa, Marco; Dell'Acqua, AndreaThe Run 2 of the Large Hadron Collider, which began in Spring 2015, offers new challenges to the Experiments with its unprecedented energy scale and high luminosity regime. To cope with the new experimental conditions, the ATLAS Experiment was upgraded during the first long shutdown of the collider, in the period 2013-2014. The most relevant change which occurred in the ATLAS Inner Detector was the installation of a fourth pixel layer, the Insertable B-Layer, at a radius of 33 mm together with a new thinner beam pipe. The Pixel Services, located between the Pixel and SCT detectors, were also modified. Owing to the radically modified ID layout, many aspects of the track reconstruction programs had to be re-optimized. In this thesis, the improvements to the tracking algorithms and the studies of the material distribution in the Inner Detector are described in detail, together with the improvements introduced in the geometry model description in simulation as well as the re-evaluation and the reduction of the systematic uncertainty on the estimate of the track reconstruction efficiency. The results of these studies were applied to the measurement of Charged-Particle Multiplicity in proton–proton collisions at a centre-of-mass energy of 13 TeV. The chargedparticle multiplicity, its dependence on transverse momentum and pseudorapidity and the dependence of the mean transverse momentum on the charged-particle multiplicity are presented for various fiducial phase spaces. The measurements are corrected for detector effects, presented as particle-level distributions and are compared to the predictions of different Monte Carlo event generators. New sets of recommended performance figures along with the related systematic uncertainties were also derived for several aspects of the ATLAS tracking, such as track reconstruction efficiency, fake rate and impact parameter resolution. These recommendations provide information on appropriate working points, i.e. track selection criteria with wellunderstood performance. They apply to physics analyses using Inner Detector tracks in Run 2 data and are important inputs for other objects based on tracks, such as jets. A simulation-based method which uses the tracking recommendations to calibrate light-jets mis-tagged as b-jets it is also presented in the context of the measurement of the crosssection of the W-boson produced in association with b-jets at 13 TeV, together with an overview of the inclusiveW-boson cross-section analysisItem Modellazione di linguaggi naturali e artificiali attraverso la scienza delle reti(2012-12-17) Bertacchini, Francesca; Bilotta, Eleonora; Pantano, PietroItem <> comunicazione scientifica nell'arte tra realtà e realtà aumentata(2014-06-05) Rinaudo, Daniela; Pantano, PietroItem <> genesi del circuito di Chua: uso del computer animation per divulgare la teoria del caos(2014-01-31) Laria, Giuseppe; Pantano, Pietro