Removing noisy mentions for distant supervision

Los métodos para Extracción de Información basados en la Supervisión a Distancia se basan en usar tuplas correctas para adquirir menciones de esas tuplas, y así entrenar un sistema tradicional de extracción de información supervisado. En este artículo analizamos las fuentes de ruido en las menciones, y exploramos métodos sencillos para filtrar menciones ruidosas. Los resultados demuestran que combinando el filtrado de tuplas por frecuencia, la información mutua y la eliminación de menciones lejos de los centroides de sus respectivas etiquetas mejora los resultados de dos modelos de extracción de información significativamente.

Flexible ruthenium oxide-activated carbon cloth composites prepared by simple electrodeposition methods

This work focuses on the preparation of flexible ruthenium oxide containing activated carbon cloth by electrodeposition. Different electrodeposition methods have been used, including chronoamperometry, chronopotentiometry and cyclic voltammetry. The electrochemical properties of the obtained materials have been measured. The results show that the potentiostatic method allows preparing composites with higher specific capacitance than the pristine activated carbon cloth. The capacitance values measured by cyclic voltammetry at 10 mV s−1 and 1 V of potential window were up to 160 and 180 F g−1. This means an improvement of 82% and 100% with respect to the capacitance of the pristine activated carbon cloth. This excellent capacitance enhancement is attributed to the small particle size (4–5 nm) and the three-dimensional nanoporous network of the ruthenium oxide film which allows reaching very high degree of oxide utilization without blocking the pore structure of the activated carbon cloth. In addition, the electrodes maintain the mechanical properties of the carbon cloth and can be useful for flexible devices.

Local convergence of quasi-Newton methods under metric regularity

We consider quasi-Newton methods for generalized equations in Banach spaces under metric regularity and give a sufficient condition for q-linear convergence. Then we show that the well-known Broyden update satisfies this sufficient condition in Hilbert spaces. We also establish various modes of q-superlinear convergence of the Broyden update under strong metric subregularity, metric regularity and strong metric regularity. In particular, we show that the Broyden update applied to a generalized equation in Hilbert spaces satisfies the Dennis–Moré condition for q-superlinear convergence. Simple numerical examples illustrate the results.

A Simple RVoG Test for PolInSAR Data

In this paper, we present a simple algorithm for assessing the validity of the RVoG model for PolInSAR-based inversion techniques. This approach makes use of two important features characterizing a homogeneous random volume over a ground surface, i.e., the independence on polarization states of wave propagation through the volume and the structure of the polarimetric interferometric coherency matrix. These two features have led to two different methods proposed in the literature for retrieving the topographic phase within natural covers, i.e., the well-known line fitting procedure and the observation of the (1, 2) element of the polarimetric interferometric coherency matrix. We show that differences between outputs from both approaches can be interpreted in terms of the PolInSAR modeling based on the Freeman-Durden concept, and this leads to the definition of a RVoG/non-RVoG test. The algorithm is tested with both indoor and airborne data over agricultural and tropical forest areas.