Feature Discretization with Relevance and Mutual Information Criteria

Feature discretization (FD) techniques often yield adequate and compact representations of the data, suitable for machine learning and pattern recognition problems. These representations usually decrease the training time, yielding higher classification accuracy while allowing for humans to better understand and visualize the data, as compared to the use of the original features. This paper proposes two new FD techniques. The first one is based on the well-known Linde-Buzo-Gray quantization algorithm, coupled with a relevance criterion, being able perform unsupervised, supervised, or semi-supervised discretization. The second technique works in supervised mode, being based on the maximization of the mutual information between each discrete feature and the class label. Our experimental results on standard benchmark datasets show that these techniques scale up to high-dimensional data, attaining in many cases better accuracy than existing unsupervised and supervised FD approaches, while using fewer discretization intervals.

The insular shelves of the Faial-Pico Ridge (Azores archipelago): a morphological record of its evolution

Shelves surrounding reefless volcanic ocean islands are formed by surf erosion of their slopes during changing sea levels. Posterosional lava flows, if abundant, can cross the coastal cliffs and fill partially or completely the accommodation space left by erosion. In this study, multibeam bathymetry, high-resolution seismic reflection profiles, and sediment samples are used to characterize the morphology of the insular shelves adjacent to Pico Island. The data show offshore fresh lava flow morphologies, as well as an irregular basement beneath shelf sedimentary bodies and reduced shelf width adjacent to older volcanic edifices in Pico. These observations suggest that these shelves have been significantly filled by volcanic progradation and can thus be classified as rejuvenated. Despite the general volcanic infilling of the shelves around Pico, most of their edges are below the depth of the Last Glacial Maximum, revealing that at least parts of the island have subsided after the shelves formed by surf erosion. Prograding lava deltas reached the shelf edge in some areas triggering small slope failures, locally decreasing the shelf width and depth of their edges. These areas can represent a significant risk for the local population; hence, their identification can be useful for hazard assessment and contribute to wiser land use planning. Shelf and subaerial geomorphology, magnetic anomalies and crustal structure data of the two islands were also interpreted to reconstruct the long-term combined onshore and offshore evolution of the Faial-Pico ridge. The subaerial emergence of this ridge is apparently older than previously thought, i.e., before approximate to 850 ka.