Results 11  20
of
102
Empirical Acquisition of Conceptual Distinctions via Dictionary Definitions
, 2004
"... This thesis discusses the automatic acquisition of conceptual distinctions using empirical methods, with an emphasis on semantic relations. The goal is to improve semantic lexicons for computational linguistics, but the work can be applied to generalpurpose knowledge bases as well. ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
This thesis discusses the automatic acquisition of conceptual distinctions using empirical methods, with an emphasis on semantic relations. The goal is to improve semantic lexicons for computational linguistics, but the work can be applied to generalpurpose knowledge bases as well.
SENSITIVITY OF INFERENCES IN FORENSIC GENETICS TO ASSUMPTIONS ABOUT FOUNDING GENES
"... Many forensic genetics problems can be handled using structured systems of discrete variables, for which Bayesian networks offer an appealing practical modeling framework, and allow inferences to be computed by probability propagation methods. However, when standard assumptions are violated—for exam ..."
Abstract

Cited by 3 (2 self)
 Add to MetaCart
Many forensic genetics problems can be handled using structured systems of discrete variables, for which Bayesian networks offer an appealing practical modeling framework, and allow inferences to be computed by probability propagation methods. However, when standard assumptions are violated—for example, when allele frequencies are unknown, there is identity by descent or the population is heterogeneous—dependence is generated among founding genes, that makes exact calculation of conditional probabilities by propagation methods less straightforward. Here we illustrate different methodologies for assessing sensitivity to assumptions about founders in forensic genetics problems. These include constrained steepest descent, linear fractional programming and representing dependence by structure. We illustrate these methods on several forensic genetics examples involving criminal identification, simple and complex disputed paternity and DNA mixtures. 1. Introduction. Forensic
Characterizing the generalized lambda distribution by Lmoments
 Computational Statistics and Data Analysis
, 2008
"... The generalized lambda distribution (GLD) is a flexible four parametric distribution with many practical applications. Lmoments of the GLD can be expressed in closed form and are good alternatives for the central moments. In this paper, we present the Lmoments of the GLD up to an arbitrary order a ..."
Abstract

Cited by 3 (0 self)
 Add to MetaCart
The generalized lambda distribution (GLD) is a flexible four parametric distribution with many practical applications. Lmoments of the GLD can be expressed in closed form and are good alternatives for the central moments. In this paper, we present the Lmoments of the GLD up to an arbitrary order and study which values of Lskewness and Lkurtosis can be achieved by the GLD. The boundaries of Lskewness and Lkurtosis are derived analytically in the symmetric case and calculated numerically in the general case. In addition, the contours of Lskewness and Lkurtosis are presented as functions of the GLD parameters. It is found that with an exception of the smallest values of Lkurtosis, the GLD covers all possible pairs of Lskewness and Lkurtosis and often there are two or more distributions that share the same Lskewness and the same Lkurtosis. We present an example that demonstrates a situation where there are four GLD members with the same Lskewness and the same Lkurtosis. The results increase our knowledge on the distributions that belong to the GLD family and can be utilized in model selection and estimation. Keywords: skewness, kurtosis, Lmoment ratio diagram, method of moments, method of Lmoments 1 1
Increasing Informativeness in Temporal Annotation
"... In this paper, we discuss some of the challenges of adequately applying a specification language to an annotation task, as embodied in a specific guideline. In particular, we discuss some issues with TimeML motivated by error analysis on annotated TLINKs in TimeBank. We introduce a document level i ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
In this paper, we discuss some of the challenges of adequately applying a specification language to an annotation task, as embodied in a specific guideline. In particular, we discuss some issues with TimeML motivated by error analysis on annotated TLINKs in TimeBank. We introduce a document level information structure we call a narrative container (NC), designed to increase informativeness and accuracy of temporal relation identification. The narrative container is the default interval containing the events being discussed in the text, when no explicit temporal anchor is given. By exploiting this notion in the creation of a new temporal annotation over TimeBank, we were able to reduce inconsistencies and increase informativeness when compared to existing TLINKs in TimeBank. 1
Perfect simulation and moment properties for the Matérn Type III Process
, 2009
"... process) is a less wellknown but for many applications more appealing or realistic model than the Matérn type I and II hard core point processes. This paper focuses on the stationary (and hence infinite) Matérn III process from a probabilistic and a stochastic geometry perspective. Briefly, given a ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
process) is a less wellknown but for many applications more appealing or realistic model than the Matérn type I and II hard core point processes. This paper focuses on the stationary (and hence infinite) Matérn III process from a probabilistic and a stochastic geometry perspective. Briefly, given a hard core parameter R> 0, the Matérn III process is obtained by a dependent thinning from a spatiotemporal Poisson process on R d ×[0, 1] with intensity λ> 0, where a Poisson point becomes a Matérn III point if the ball of radius R centered at the point does not contain an earlier Matérn III point. Using a construction of Matérn III that creates various ‘generations ’ of points, a perfect simulation algorithm for the infinite Matérn III process within a bounded region is developed. It is shown that the log expected number of points that must be examined is bounded above by a linear function which is easily calculated. This result is quite general, which is illustrated by an extension of the basic Matérn III process to allow random radii or more generally to replace balls with random sets, and also to allow spatial inhomogeneity. The perfect simulation algorithm is used to provide Monte Carlo estimates of the packing density of Matérn III, which can be much higher than for Matérn I or II, and increases to the jamming limit of the random sequential adsorption model as λ → ∞.
Effects of breeding success, mate fidelity and senescence on breeding dispersal of male and female bluefooted boobies
 J Anim
, 2007
"... 1. Understanding the effects of individual and population factors on variation in breeding dispersal (the movement of individuals between successive breeding sites) is key to identifying the strategies behind breeders ’ movements. Dispersal is often influenced by multiple factors and these can be co ..."
Abstract

Cited by 1 (1 self)
 Add to MetaCart
1. Understanding the effects of individual and population factors on variation in breeding dispersal (the movement of individuals between successive breeding sites) is key to identifying the strategies behind breeders ’ movements. Dispersal is often influenced by multiple factors and these can be confounded with each other. We used 13 years of data on the locations, mates, breeding success and ages of individuals to tease apart the factors influencing breeding dispersal in a colonially breeding longlived seabird, the bluefooted booby Sula nebouxii. 2. Breeding dispersal varied among and within years. Males dispersed further in years of higher population density, and late breeding males and females dispersed further than early breeders. This temporal variation related to changes in competition for territory was taken into account in all tests of individual factors influencing breeding dispersal. 3. Individuals that retained their mates from the previous year dispersed shorter distances than those that changed their mates. 4. The effect of previous breeding success depended on mate fidelity. Unsuccessful breeding induced greater dispersal in birds that changed their mates but not in birds that
Hyperspectral Analysis of Soil Nitrogen, Carbon, Carbonate, and Organic Matter Using Regression Trees
, 2012
"... sensors ..."
Notes ShortTerm Load Forecasting with Neural Network Ensembles: A Comparative Study
"... Load Forecasting plays a critical role in the management, scheduling and dispatching operations ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
Load Forecasting plays a critical role in the management, scheduling and dispatching operations
Similarities and Differences in Lipidomics Profiles among Healthy Monozygotic Twin Pairs
"... Differences in genetic background and/or environmental exposure among individuals are expected to give rise to differences in measurable characteristics, or phenotypes. Consequently, genetic resemblance and similarities in environment should manifest as similarities in phenotypes. The metabolome ref ..."
Abstract
 Add to MetaCart
Differences in genetic background and/or environmental exposure among individuals are expected to give rise to differences in measurable characteristics, or phenotypes. Consequently, genetic resemblance and similarities in environment should manifest as similarities in phenotypes. The metabolome reflects many of the system properties, and is therefore an important part of the phenotype. Nevertheless, it has not yet been examined to what extent individuals sharing part of their genome and/or environment indeed have similar metabolomes. Here we present the results of hierarchical clustering of blood plasma lipid profile data obtained by liquid chromatography–mass spectrometry from 23 healthy, 18yearold twin pairs, of which 21 pairs were monozygotic, and 8 of their siblings. For 13 monozygotic twin pairs, withinpair similarities in relative concentrations of the detected lipids were indeed larger than the similarities with any other study participant. We demonstrate such high coclustering to be unexpected on basis of chance. The similarities between dizygotic twins and between nontwin siblings, as well as between nonfamilial participants, were less pronounced. In a number of twin pairs, withinpair dissimilarity of lipid profiles
E 2008, by the American Society of Limnology and Oceanography, Inc. A framework for developing ecosystemspecific nutrient criteria: Integrating biological thresholds with predictive modeling
"... We present a novel ecosystemspecific framework for developing nutrient criteria from biological thresholds and predictive modeling (BTPM) and an application of this framework to lakes in Michigan, U.S. The four main components for the BTPM framework are: (1) to predict each ecosystem’s ‘expected nu ..."
Abstract
 Add to MetaCart
We present a novel ecosystemspecific framework for developing nutrient criteria from biological thresholds and predictive modeling (BTPM) and an application of this framework to lakes in Michigan, U.S. The four main components for the BTPM framework are: (1) to predict each ecosystem’s ‘expected nutrient concentration ’ in the absence of human effects using a predictive model, (2) to identify important biological thresholds along a nutrient gradient (i.e., biological [BIO] benchmarks), (3) to determine each ecosystem’s current nutrient concentration, and (4) to use the above information to derive a nutrient criterion for each ecosystem using the BTPM algorithm. The BTPM framework is extremely flexible in that it can be applied to any aquatic ecosystem type or nutrient and the four components can be implemented in a variety of ways. Our BTPM framework has two additional features: it recognizes that prior to human disturbance, ecosystems varied in their natural nutrient concentrations, and it incorporates risk into the decisionmaking process. In the simplest scheme, a nutrient criterion is set at a BIO benchmark greater than the expected nutrient concentration. However, to protect ecosystems more conservatively, a criterion is set at current lake nutrient concentrations if current is less than the BIO benchmark. In our application of the BTPM framework, we developed total phosphorus (TP) criteria for a diverse set of 374 lakes in MI. The expected lake TP concentrations in the absence of human effects ranged from 3 mg L 21 to 24 mg L 21, suggesting that a single criterion approach would not be