Investigando a combinação de técnicas de aprendizado semissupervisionado e classificação hierárquica multirrótulo

Data classification is a task with high applicability in a lot of areas. Most methods for treating classification problems found in the literature dealing with single-label or traditional problems. In recent years has been identified a series of classification tasks in which the samples can be labeled at more than one class simultaneously (multi-label classification). Additionally, these classes can be hierarchically organized (hierarchical classification and hierarchical multi-label classification). On the other hand, we have also studied a new category of learning, called semi-supervised learning, combining labeled data (supervised learning) and non-labeled data (unsupervised learning) during the training phase, thus reducing the need for a large amount of labeled data when only a small set of labeled samples is available. Thus, since both the techniques of multi-label and hierarchical multi-label classification as semi-supervised learning has shown favorable results with its use, this work is proposed and used to apply semi-supervised learning in hierarchical multi-label classication tasks, so eciently take advantage of the main advantages of the two areas. An experimental analysis of the proposed methods found that the use of semi-supervised learning in hierarchical multi-label methods presented satisfactory results, since the two approaches were statistically similar results

Análise multitemporal dos elementos geoambientais da região da Ponta do Tubarão, área de influência dos campos petrolíferos de Macau e Serra, município de Macau/RN

The study area is inserted in Ponta do Tubarão region, Macau City, setentrional littoral of Rio Grande do Norte State, composed of Tertiary and Quatemary sedimentary rocks and sediments. This region is characterized for the intense action of the coastal processes, causing the morphologic instability in part of the area, beyond the interference of human activities, as the Petroliferous Industrial Polo, salt companies and shrimp farms. This justifies the integration of multidisciplinary and multitemporal detailed scientific studies dealing with the evaluation of the changing behavior of this coastal environment by geoenvironments elements characterization, identifying protected and recuperation areas, mainly those under socioeconomic intervention. The main objective was the coastal monitoring using geoprocessing techniques to prepare thematic maps useful for oil spilling environment risk areas survey. The methodology was based on multitemporal interpretation of remote sensing images and field checking, integrated in a Geographical Information System (GIS). The Geologic, Geomorphologic, Vegetation, Soil and Land Use maps were prepared, and later on they allowed the generation of the Natural Vulnerability and Environmental Vulnerability maps. These maps had been classified in accordance with vulnerability degrees: very low, low, medi um, high and very high. Beyond these maps the GIS allowed the analysis of the shoreline evolution for 10 distinct dates, using Landsat 5 TM and 7 ETM+ and SPOT-HRVIR images. This analysis made possible the attendance of the coastal morphodynamic evolution, where the results had been represented by areasof erosion and accretion (or deposition) of sediments, pointing critical areas under erosive process to the petroliferous industry (Macau and Serra fields). The GIS also provided to prepare the Environmental Sensitivity Maps of Oil Spill (SAO Maps) in operational scale (1: 10.000), according to the norms ofthe Ministério do Meio Ambiente (MMA 2002). The SAO Map in operational scale was based on IKONOS images mosaic where the ESI (Environmental Sensitivity Index) was represented according with two tides phases of theregion. Therewere recognizedfiveESI (3, 4,7,9, 1O) for the low tide; to the high tide the ESI number increased to seven (3, 4, 5, 7, 8, 9, 10). All these information are necessary to the decisions making about oi! spill and its derivatives containment. These techniques application makes possible the optimization and implantation ofnew socioeconomics activities of low environmental impact, indicates areas for better productivity and security exploration, and benefits local communities with fauna and flora preservation. The development of these activities is inserted in the scope of Monitoramento Ambiental de Áreas de Risco a Derrames de Petróleo e Seus Derivados Cooperation Project (Rede 05/01 - PETRORISCO, FINEP/CTPETRO/PETROBRAS) of multidisciplinary and interinstitucional characteristics dealing with subjects involving the environmental monitoring and the petroliferous activity