DSpace - Universidade do Porto
(66,537 recursos)
Este repositório institucional visa coleccionar, preservar e disponibilizar na Internet a produção intelectual da comunidade académica da U.PORTO. Apenas as publicações em texto integral e de acesso livre (sob as condições de licença pública Creative Commons)fazem parte deste repositório. Informações adicionais sobre outras publicações encontram-se em www.up.pt.

FCUP - Artigo em Acta de Conferência Internacional

Mostrando recursos 1 - 15 de 15

1.
A pipelined data-parallel algorithm for ILP - fonseca, nuno a.; silva, fernando; costa, vitor santos; camacho, rui
The amount of data collected and stored in databases is growing considerably for almost all areas of human activity. Processing this amount of data is very expensive, both humanly and computationally. This justifies the increased interest both on the automatic discovery of useful knowledge from databases, and on using parallel processing for this task. Multi Relational Data Mining (MRDM) techniques, such as Inductive Logic Programming (ILP), can learn rules from relational databases consisting of multiple tables. However current ILP systems are designed to run in main memory and can have long running times. We propose a pipelined data-parallel algorithm for...

2.
A pipelined data-parallel algorithm for ILP - fonseca, nuno a.; silva, fernando; costa, vitor santos; camacho, rui
The amount of data collected and stored in databases is growing considerably for almost all areas of human activity. Processing this amount of data is very expensive, both humanly and computationally. This justifies the increased interest both on the automatic discovery of useful knowledge from databases, and on using parallel processing for this task. Multi Relational Data Mining (MRDM) techniques, such as Inductive Logic Programming (ILP), can learn rules from relational databases consisting of multiple tables. However current ILP systems are designed to run in main memory and can have long running times. We propose a pipelined data-parallel algorithm for...

3.
Time Dependent Clustering of Time Series - Joaquim Costa; Isabel Silva; M. Eduarda Silva
In this work we consider the problem of clustering time series. Contrary to other works on this topic, our main concern is to let the most important observations, for instance the most recent, have a larger weight on the analysis. This is done by defining similarities measures between two time series, based on Pearsons correlation coefficient, which uses the notion of weighted mean and weighted covariance, where the weights increase monotonically with the time. We use these measures, which are metrics between time series, as a similarity or dissimilarity index between the $n$ time series to be clustered. We apply...

7.
Probabilistic surface change detection and measurement from digital aerial stereo images
We propose a new method to measure changes in terrain topography from two optical stereo image pairs acquired at different
dates. The main novelty is in the ability of computing the spatial distribution of uncertainty, thanks to stochastic modeling and
probabilistic inference. Thus, scientists will have access to quantitative error estimates of local surface variation, so they can
check the statistical significance of elevation changes, and make, where changes have occurred, consistent measurements of
volume or shape evolution. The main application area is geomorphology, as the method can help study phenomena such as
coastal cliff erosion, sand dune displacement and various transport mechanisms through the...

8.
The use of similarity images on multi-sensor automatic image registration - H. Gonçalves; J.A. Gonçalves; L. Corte-Real
Automatic image registration (AIR) is still a present challenge regarding remote sensing applications. Although several methodshave been proposed in the last few years, geometric correction is often a time and effort consuming manual task. The only AIRmethod which is commonly used is the correlation-based template matching method. It usually consists on considering a windowfrom one image and passing it throughout the other, looking for a maximum of correlation, which may be associated to thedisplacement between the two images. This approach leads sometimes (for example with multi-sensor image registration) to lowcorrelation coefficient values, which do not give sufficient confidence to associate...

11.
MONITORING OF BEACHES AND SAND DUNES USING DIGITAL AERIAL
PHOTOGRAPHY WITH DIRECT GEOREFERENCING
Beaches in the Portuguese western coast suffer sudden changes due to the action of the sea during winter. Frequent
acquisition of digital surface models (DSM) is needed in order to assess volumetric changes. This paper presents a
monitoring program carried in an area of 15 km near the city of Porto. Digital aerial photography is acquired twice
a year, using a Zeiss-Intergraph DMC camera, with a spatial resolution of 10 cm, by a local company. Images are
provided with exterior orientation provided by direct georeferencing equipment. An assessment of the exterior
orientation parameters provided was done using field survey points. It could be concluded that...

12.
Deployment and evaluation of a usage based collaborative filtering recommendation system with blacklists - Luís Lemos; Alípio M. Jorge; José Paulo Leal
Social web sites, like Palco Principal (PP), have considerably sizeddatabases with the most varied types of information from user subscriptioninformation to user interactions. All that information is susceptible of beingprocessed using data mining algorithms in order to extract knowledge relevantfor the business. This document describes the implementation and evaluation ofa system that uses data mining techniques, and that has been developed with thespecific intent of generating recommendations, namely musicrecommendations, to users of the PP web site. The implemented system usescollaborative filtering techniques to recommend music additions to a user#8217;splaylist. The impact of the deployed system has been evaluated online.

13.
Kostia Beidars contribution to module and ring theory - Christian Lomp; Robert Wisbauer
At the beginning of his mathematical career Kostia Beidar was working on rings with polynomial identities and primeness conditions for rings. By Posner#8217;s theorem the two-sided quotient ring of a prime PI-ring is a finite matrix ring over some field. This result was extended by Martindale to rings with generalised polynomial identities by the construction of the central closure of a prime ring. Kostia was working extensively in this setting and made crucial contributions to the understanding of the theory.While his contribution to general PI theory will be outlinedelsewhere we want to sketch here his work on prime rings and...

14.
Comparability, distributivity and non-commutative #966;-rings. - Alveri SantAna; Christian Lomp
The purpose of this paper is to develop the theory of non-commutative #966;-rings using concepts from distributive rings and rings with comparability. We recover the main statements for commutative #966;-rings as obtained by Anderson and Badawi.

15.
The splitting of the dual Goldie torsion theory
The splitting of the Goldie (or singular) torsion theory has been extensively studied. Here we determine an appropriate dual Goldie torsion theory, discuss its splitting and answer in the negative a question proposed by Ozcan and Harmanci as to whether the splitting of the dual Goldie torsion theory implies the ring to be quasi-Frobenius.