Extração semi-automática do eixo de rodovia em imagens de média e alta resolução usando programação dinâmica

Pós-graduação em Ciências Cartográficas - FCT

Entomofauna associada a goiabeira Psidium guajava L. em pomares experimentais comerciais em Vista Alegre do Alto – SP e semi-orgânicos em Pindorama – SP

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Defaunação e efeitos-cascata sobre a diversidade vegetal em uma ilha semi-defaunada na Floresta Atlântica

Pós-graduação em Ciências Biológicas (Zoologia) - IBRC

Caracterização da variabilidade morfoagronômica de maracujazeiro (Passiflora cincinnata Mast.) no semi-árido brasileiro

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Manejo da floração de mangueiras no Semi-árido do nordeste brasileiro com inibidores da síntese de giberelinas

Pós-graduação em Agronomia (Horticultura) - FCA

Semi-supervised learning with concept drift using particle dynamics applied to network intrusion detection data

Concept drift, which refers to non stationary learning problems over time, has increasing importance in machine learning and data mining. Many concept drift applications require fast response, which means an algorithm must always be (re)trained with the latest available data. But the process of data labeling is usually expensive and/or time consuming when compared to acquisition of unlabeled data, thus usually only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are based on assumptions that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenging task in machine learning. Recently, a particle competition and cooperation approach has been developed to realize graph-based semi-supervised learning from static data. We have extend that approach to handle data streams and concept drift. The result is a passive algorithm which uses a single classifier approach, naturally adapted to concept changes without any explicit drift detection mechanism. It has built-in mechanisms that provide a natural way of learning from new data, gradually "forgetting" older knowledge as older data items are no longer useful for the classification of newer data items. The proposed algorithm is applied to the KDD Cup 1999 Data of network intrusion, showing its effectiveness.

Semi-supervised dimensionality reduction based on partial least squares for visual analysis of high dimensional data

Dimensionality reduction is employed for visual data analysis as a way to obtaining reduced spaces for high dimensional data or to mapping data directly into 2D or 3D spaces. Although techniques have evolved to improve data segregation on reduced or visual spaces, they have limited capabilities for adjusting the results according to user's knowledge. In this paper, we propose a novel approach to handling both dimensionality reduction and visualization of high dimensional data, taking into account user's input. It employs Partial Least Squares (PLS), a statistical tool to perform retrieval of latent spaces focusing on the discriminability of the data. The method employs a training set for building a highly precise model that can then be applied to a much larger data set very effectively. The reduced data set can be exhibited using various existing visualization techniques. The training data is important to code user's knowledge into the loop. However, this work also devises a strategy for calculating PLS reduced spaces when no training data is available. The approach produces increasingly precise visual mappings as the user feeds back his or her knowledge and is capable of working with small and unbalanced training sets.

Analisi e progettazione di un sistema di link discovery semi-automatico nell'ambito dei Linked Open Data

Lo scopo di questo elaborato è di analizzare e progettare un sistema in grado di supportare la definizione dei dati nel formato utilizzato per definire in modo formale la semantica dei dati, ma soprattutto nella complessa e innovativa attività di link discovery. Una attività molto potente che, tramite gli strumenti e le regole del Web Semantico (chiamato anche Web of Data), permette data una base di conoscenza sorgente ed altre basi di conoscenza esterne e distribuite nel Web, di interconnettere i dati della base di conoscenza sorgente a quelli esterni sulla base di complessi algoritmi di interlinking. Questi algoritmi fanno si che i concetti espressi sulla base di dati sorgente ed esterne vengano interconnessi esprimendo la semantica del collegamento ed in base a dei complessi criteri di confronto definiti nel suddetto algoritmo. Tramite questa attività si è in grado quindi di aumentare notevolmente la conoscenza della base di conoscenza sorgente, se poi tutte le basi di conoscenza presenti nel Web of Data seguissero questo procedimento, la conoscenza definita aumenterebbe fino a livelli che sono limitati solo dalla immensa vastità del Web, dando una potenza di elaborazione dei dati senza eguali. Per mezzo di questo sistema si ha l’ambizioso obiettivo di fornire uno strumento che permetta di aumentare sensibilmente la presenza dei Linked Open Data principalmente sul territorio nazionale ma anche su quello internazionale, a supporto di enti pubblici e privati che tramite questo sistema hanno la possibilità di aprire nuovi scenari di business e di utilizzo dei dati, dando una potenza al dato che attualmente è solo immaginabile.

Data integration, real-time synchronization and RESTful Web Service

Analisi e sviluppo di procedure di importazione dati per un integratore di annunci immobiliari dedicato alla vendita di soggiorni turistici in case vacanza. Il documento tratta inoltre l'implementazione di un Web Service conforme all'architettura RESTful per l'accesso e l'esportazione dei dati a soggetti terzi autorizzati tramite Digest Authentication.

Measurement of the TeV atmospheric muon charge ratio with the complete OPERA data set

The OPERA detector, designed to search for νμ → ντ oscillations in the CNGS beam, is located in the underground Gran Sasso laboratory, a privileged location to study TeV-scale cosmic rays. For the analysis here presented, the detector was used to measure the atmospheric muon charge ratio in the TeV region. OPERA collected chargeseparated cosmic ray data between 2008 and 2012. More than 3 million atmospheric muon events were detected and reconstructed, among which about 110000 multiple muon bundles. The charge ratio Rμ ≡ Nμ+/Nμ− was measured separately for single and for multiple muon events. The analysis exploited the inversion of the magnet polarity which was performed on purpose during the 2012 Run. The combination of the two data sets with opposite magnet polarities allowedminimizing systematic uncertainties and reaching an accurate determination of the muon charge ratio. Data were fitted to obtain relevant parameters on the composition of primary cosmic rays and the associated kaon production in the forward fragmentation region. In the surface energy range 1–20 TeV investigated by OPERA, Rμ is well described by a parametric model including only pion and kaon contributions to themuon flux, showing no significant contribution of the prompt component. The energy independence supports the validity of Feynman scaling in the fragmentation region up to 200 TeV/nucleon primary energy.