MongoDB: analisi e prototipazione su applicazioni di social business intelligence

L’ultimo decennio ha visto un radicale cambiamento del mercato informatico, con la nascita di un numero sempre maggiore di applicazioni rivolte all’interazione tra utenti. In particolar modo, l’avvento dei social network ha incrementato notevolmente le possibilità di creare e condividere contenuti sul web, generando volumi di dati sempre maggiori, nell’ordine di petabyte e superiori. La gestione di tali quantità di dati ha portato alla nascita di soluzioni non relazionali appositamente progettate, dette NoSQL. Lo scopo di questo documento è quello di illustrare come i sistemi NoSQL, nello specifico caso di MongoDB, cerchino di sopperire alle difficoltà d’utilizzo dei database relazionali in un contesto largamente distribuito. Effettuata l'analisi delle principali funzionalità messe a disposizione da MongoDB, si illustreranno le caratteristiche di un prototipo di applicazione appositamente progettato che sfrutti una capacità peculiare di MongoDB quale la ricerca full-text. In ultima analisi si fornirà uno studio delle prestazioni di tale soluzione in un ambiente basato su cluster, evidenziandone il guadagno prestazionale.

Competence: intelligence in sheep's clothing? Culture, representations, and cognitive performance

The first aim of this thesis was to contribute to the understanding of how cultural capital (Bourdieu, 1983/1986) affects students achievements and performances. We specifically claimed that the effect of cultural capital is at least partly explained by the positioning students take towards the principles they use to attribute competence and intelligence. The testing of these hypothesis have been framed within the social representations theory, specifically in the formulation of the Lemanic school approach (Doise, 1986).

Technologies for Ambient intelligence: from Smart Objects to Sensor Networks

n the last few years, the vision of our connected and intelligent information society has evolved to embrace novel technological and research trends. The diffusion of ubiquitous mobile connectivity and advanced handheld portable devices, amplified the importance of the Internet as the communication backbone for the fruition of services and data. The diffusion of mobile and pervasive computing devices, featuring advanced sensing technologies and processing capabilities, triggered the adoption of innovative interaction paradigms: touch responsive surfaces, tangible interfaces and gesture or voice recognition are finally entering our homes and workplaces. We are experiencing the proliferation of smart objects and sensor networks, embedded in our daily living and interconnected through the Internet. This ubiquitous network of always available interconnected devices is enabling new applications and services, ranging from enhancements to home and office environments, to remote healthcare assistance and the birth of a smart environment. This work will present some evolutions in the hardware and software development of embedded systems and sensor networks. Different hardware solutions will be introduced, ranging from smart objects for interaction to advanced inertial sensor nodes for motion tracking, focusing on system-level design. They will be accompanied by the study of innovative data processing algorithms developed and optimized to run on-board of the embedded devices. Gesture recognition, orientation estimation and data reconstruction techniques for sensor networks will be introduced and implemented, with the goal to maximize the tradeoff between performance and energy efficiency. Experimental results will provide an evaluation of the accuracy of the presented methods and validate the efficiency of the proposed embedded systems.

Inverse Static Analysis of Massive Parallel Arrays of Three-State Actuators via Artificial Intelligence

Massive parallel robots (MPRs) driven by discrete actuators are force regulated robots that undergo continuous motions despite being commanded through a finite number of states only. Designing a real-time control of such systems requires fast and efficient methods for solving their inverse static analysis (ISA), which is a challenging problem and the subject of this thesis. In particular, five Artificial intelligence methods are proposed to investigate the on-line computation and the generalization error of ISA problem of a class of MPRs featuring three-state force actuators and one degree of revolute motion.

Trait emotional intelligence: Evidence from italian adolescents and adults

Emotional intelligence (EI) represents an attribute of contemporary attractiveness for the scientific psychology community. Of particular interest for the present thesis are the conundrum related to the representation of this construct conceptualized as a trait (i.e., trait EI), which are in turn reflected in the current lack of agreement upon its constituent elements, posing significant challenges to research and clinical progress. Trait EI is defined as an umbrella personality-alike construct reflecting emotion-related dispositions and self-perceptions. The Trait Emotional Intelligence Questionnaire (TEIQue) was chosen as main measure, given its strong theoretical and psychometrical basis, including superior predictive validity when compared to other trait EI measures. Studies 1 and 2 aimed at validating the Italian 153-items forms of the TEIQue devoted to adolescents and adults. Analyses were done to investigate the structure of the questionnaire, its internal consistencies and gender differences at the facets, factor, and global level of both versions. Despite some low reliabilities, results from Studies 1 and 2 confirm the four-factor structure of the TEIQue. Study 3 investigated the utility of trait EI in a sample of adolescents over internalizing conditions (i.e., symptoms of anxiety and depression) and academic performance (grades at math and Italian language/literacy). Beyond trait EI, concurrent effects of demographic variables, higher order personality dimensions and non-verbal cognitive ability were controlled for. Study 4a and Study 4b addressed analogue research questions, through a meta-analysis and new data in on adults. In the latter case, effects of demographics, emotion regulation strategies, and the Big Five were controlled. Overall, these studies showed the incremental utility of the TEIQue in different domains beyond relevant predictors. Analyses performed at the level of the four-TEIQue factors consistently indicated that its predictive effects were mainly due to the factor Well-Being. Findings are discussed with reference to potential implication for theory and practice.

Studio e implementazione di una soluzione di business intelligence a supporto dell'analisi economico finanziaria

Progetto di tesi svolto in azienda. Studio dei principali concetti di Business Intelligence (BI) e degli strumenti per la BI. Confronto tra i principali vendor nel mercato dell'analisi dei dati e della Business Intelligence. Studio e reigegnerizzazione di un modello per l'analisi economico finanziaria dei fornitori/clienti di un'azienda. Realizzazione di un prototipo del modello utilizzando un nuovo strumento per la reportistica: Tableau. Il prototipo si basa su dati economici finanziari estratti da banche dati online e forniti dall'azienda cliente. Implementazione finale del database e di un flusso automatico per la riclassificazione dei dati di bilancio.

Analisi e sperimentazione del DBMS noSQL MongoDB: il caso di studio della Social Business Intelligence

Il primo capitolo prevede un’introduzione sul modello relazionale e sulle difficoltà che possono nascere nel tentativo di conformare le esigenze attuali di applicazioni ed utenti ai vincoli da esso imposti per lasciare poi spazio ad un’ampia descrizione del movimento NoSQL e delle tecnologie che ne fanno parte; il secondo capitolo sarà invece dedicato a MongoDB, alla presentazione delle sue caratteristiche e peculiarità, cercando di fornirne un quadro apprezzabile ed approfondito seppure non completo e del tutto esaustivo; infine nel terzo ed ultimo capitolo verrà approfondito il tema della ricerca di testo in MongoDB e verranno presentati e discussi i risultati ottenuti dai nostri test.

Progettazione e prototipazione di un sistema di Social Business Intelligence con ElasticSearch

Il presente elaborato ha come oggetto la progettazione e lo sviluppo di una soluzione Elasticsearch come piattaforma di analisi in un contesto di Social Business Intelligence. L’elaborato si inserisce all’interno di un progetto del Business Intelligence Group dell’Università di Bologna, incentrato sul monitoraggio delle discussioni online sul tema politico nel periodo delle elezioni europee del 2014.

Progettazione e prototipazione di un sistema di Social Business Intelligence con Hadoop Impala

Il presente elaborato ha come oggetto l’analisi delle prestazioni e il porting di un sistema di SBI sulla distribuzione Hadoop di Cloudera. Nello specifico è stato fatto un porting dei dati del progetto WebPolEU. Successivamente si sono confrontate le prestazioni del query engine Impala con quelle di ElasticSearch che, diversamente da Oracle, sfrutta la stessa componente hardware (cluster).

Posttraumatic stress disorder and soluble cellular adhesion molecules at rest and in response to a trauma-specific interview in patients after myocardial infarction

Posttraumatic stress disorder (PTSD) and circulating cellular adhesion molecules (CAMs) predict cardiovascular risk. We hypothesized a positive relationship between PTSD caused by myocardial infarction (MI) and soluble CAMs. We enrolled 22 post-MI patients with interviewer-rated PTSD and 22 post-MI patients with no PTSD. At 32±6months after index MI, all patients were re-scheduled to undergo the Clinician-Administered PTSD Scale (CAPS) interview and had blood collected to assess soluble CAMs at rest and after the CAPS interview. Relative to patients with no PTSD, those with PTSD had significantly higher levels of soluble vascular cellular adhesion molecule (sVCAM)-1 and intercellular adhesion molecule (sICAM)-1 at rest and, controlling for resting CAM levels, significantly higher sVCAM-1 and sICAM-1 after the interview. Greater severity of PTSD predicted significantly higher resting levels of sVCAM-1 and soluble P-selectin in patients with PTSD. At follow-up, patients with persistent PTSD (n=15) and those who had remitted (n=7) did not significantly differ in CAM levels at rest and after the interview; however, both these groups had significantly higher sVCAM-1 and sICAM-1 at rest and also after the interview compared to patients with no PTSD. Elevated levels of circulating CAMs might help explain the psychophysiologic link of PTSD with cardiovascular risk.

Momentary stress moderates procoagulant reactivity to a trauma-specific interview in patients with posttraumatic stress disorder caused by myocardial infarction

Hypercoagulability of the blood might partially explain the increased cardiovascular disease risk in posttraumatic stress disorder (PTSD) and is also triggered by anticipatory stress. We hypothesized exaggerated procoagulant reactivity in patients with PTSD in response to a trauma-specific interview that would be moderated by momentary stress levels. We examined 23 patients with interviewer-diagnosed PTSD caused by myocardial infarction (MI) and 21 post-MI patients without PTSD. A second diagnostic (i.e., trauma-specific) interview to assess posttraumatic stress severity was performed after a median follow-up of 26 months (range 12-36). Before that interview patients rated levels of momentary stress (Likert scale 0-10) and had blood collected before and after the interview. The interaction between PTSD diagnostic status at study entry and level of momentary stress before the follow-up interview predicted reactivity of fibrinogen (P=0.036) and d-dimer (P=0.002) to the PTSD interview. Among patients with high momentary stress levels, PTSD patients had greater fibrinogen (P=0.023) and d-dimer (P=0.035) reactivity than non-PTSD patients. Among patients with low momentary stress levels, PTSD patients had less d-dimer reactivity than non-PTSD patients (P=0.024); fibrinogen reactivity did not significantly differ between groups. Momentary stress levels, but not severity of posttraumatic stress, correlated with d-dimer reactivity in PTSD patients (r=0.46, P=0.029). We conclude that momentary stress levels moderated the relationship between PTSD and procoagulant reactivity to a trauma-specific interview. Procoagulant reactivity in post-MI patients with PTSD confronted with their traumatically experienced MI was observed if patients perceived high levels of momentary stress before the interview.