Search results for "boost"

Domain separation for efficient adaptive active learning

2011

This paper proposes a procedure aimed at efficiently adapting a classifier trained on a source image to a similar target image. The adaptation is carried out through active queries in the target domain following a strategy particularly designed for the case where class distributions have shifted between the two images. We first suggest a pre-selection of candidate pixels issued from the target image by keeping only those samples appearing to be lying in a region of the input space not yet covered by the existing ground truth (source domain pixels). Then, exploiting a classifier integrating instance weights, active queries are performed on the target image. As the inclusion to the training s…

Kinetics of Anti-Hepatitis B Surface Antigen Titers in Nurse Students after a Two-Year Follow-Up

2020

Infection caused by hepatitis B virus (HBV) can be prevented through a safe and effective vaccine. This study analysed the kinetics of serum antibodies against hepatitis B surface antigen (HBsAg) (anti-HBs) titers in relation to previous vaccine boosters in Italian nursing students who were followed up for two years. Serum anti-HBs titers were evaluated at the first visit, after vaccine booster (if required) and at visit after two years. Overall, 483 students (mean age = 21.7 years

Searches for beyond the Standard Model physics with boosted topologies in the ATLAS experiment using the Grid-based Tier-3 facility at IFIC-Valencia

2013

Tesis leída en la Universidad de Valencia, Departamento de Física Atóomica, Molecular y Nuclear, para optar al grado de Doctor en Física.-- 181 páginas

Prediction of BOD5 content of the inflow to the treatment plant using different methods of black box - the case study

2020

The publication presents the possibility of modeling in a 1 d advance of the content of organic compounds in the influent wastewater to the treatment plant, where the content of these compounds is determined by both the biochemical and chemical oxygen demand. To predict the quality of the wastewater at the inflow a set of indicators where used to make measurements on a daily basis. In order to develop statistical models 3 methods where used, namely: multivariate adaptive regression splines (MARS), boosted trees (BT), and genetic programming (GP). The carried-out calculations showed that, to calculate the BOD5 there can only be used models developed on the basis of the value of daily wastewa…

A comparison of ensemble algorithms for item-weighted Label Ranking

2023

Label Ranking (LR) is a non-standard supervised classification method with the aim of ranking a finite collection of labels according to a set of predictor variables. Traditional LR models assume indifference among alternatives. However, misassigning the ranking position of a highly relevant label is frequently regarded as more severe than failing to predict a trivial label. Moreover, switching two similar alternatives should be considered less severe than switching two different ones. Therefore, efficient LR classifiers should be able to take into account the similarities and individual weights of the items to be ranked. The contribution of this paper is to formulate and compare flexible i…

OPTIMIZING STOCHASTIC SUSCEPTIBILITY MODELLING FOR DEBRIS FLOW LANDSLIDES: MODEL EXPORTATION, STATISTICAL TECHNIQUES COMPARISON AND USE OF REMOTE SEN…

Il presente lavoro di ricerca è stato sviluppato al fine di approfondire approcci metodologici nell'ambito della sucscettibilità da frana. In particolare, il tema centrale della ricerca è rappresentato dal tema specifico dell'esportazione spaziale di modelli di suscettibilità nell'area mediterranea. All'interno del topic specifico dell'esportazione di modelli predittivi spaziali sono state approfondite tematiche relative all'utilizzo di differenti algoritmi o di differenti sorgenti, derivate da DEM o da coperture satellitari. The present work has been developed in order to enhance current methodological approaches within the big picture of the landslide susceptibility. In particular, the ce…

Las funciones interactivas del marcador español ‘¿no?’ Las fronteras entre la atenuación y la protección de la imagen

2020

In this paper the functions of the Spanish discourse marker ?no? are analysed from a pragmatic and an interactive perspective. Specifically, we explore the values of ?no? taking the pragmatic phenomena of mitigation and boosting, as well as the notion of affiliation as described in conversation analysis. The previous literature devoted to the study of this linguistic form has consistently identified its uses as a confirmation request or a phatic device (Fuentes, 1990, 2009; Santos Rio, 2003; Garcia Vizcaino, 2005; Montanez, 2008, 2015; Rodriguez Munoz, 2009; Moccero, 2010; Santana, 2017). This work, however, analyses how the mitigating uses interact and share features with neighbouring cate…

The Engineering of a Compression Boosting Library: Theory vs Practice in BWT Compression

2006

Data Compression is one of the most challenging arenas both for algorithm design and engineering. This is particularly true for Burrows and Wheeler Compression a technique that is important in itself and for the design of compressed indexes. There has been considerable debate on how to design and engineer compression algorithms based on the BWT paradigm. In particular, Move-to-Front Encoding is generally believed to be an "inefficient " part of the Burrows-Wheeler compression process. However, only recently two theoretically superior alternatives to Move-to-Front have been proposed, namely Compression Boosting and Wavelet Trees. The main contribution of this paper is to provide the first ex…

From First Principles to the Burrows and Wheeler Transform and Beyond, via Combinatorial Optimization

2007

AbstractWe introduce a combinatorial optimization framework that naturally induces a class of optimal word permutations with respect to a suitably defined cost function taking into account various measures of relatedness between words. The Burrows and Wheeler transform (bwt) (cf. [M. Burrows, D. Wheeler, A block sorting lossless data compression algorithm, Technical Report 124, Digital Equipment Corporation, 1994]), and its analog for labelled trees (cf. [P. Ferragina, F. Luccio, G. Manzini, S. Muthukrishnan, Structuring labeled trees for optimal succinctness, and beyond, in: Proc. of the 45th Annual IEEE Symposium on Foundations of Computer Science, 2005, pp. 198–207]), are special cases i…

Can an unglamorous non-event affect prices? The role of newspapers

2016

AbstractOur paper offers evidence that the print media can affect stock prices by covering public information. After price-to-book value figures of Italian listed shares were first published on the major national financial newspaper, the prices of value stocks did, on average, show a positive reaction. The price reaction was limited to small caps stocks and disappeared within three weeks. Over the period of analysis, we could not find any abnormal behaviour of the returns of small and value stocks on other European markets. These findings support the view that newspapers play a role in disseminating information to small investors and grabbing their attention, even if news are continuously r…