Experiments on identification of argumentative sentences

The main purpose of this study is to evaluate the best set of features that automatically enables the identification of argumentative sentences from unstructured text. As corpus, we use case laws from the European Court of Human Rights (ECHR). Three kinds of experiments are conducted: Basic Experiments, Multi Feature Experiments and Tree Kernel Experiments. These experiments are basically categorized according to the type of features available in the corpus. The features are extracted from the corpus and Support Vector Machine (SVM) and Random Forest are the used as Machine learning algorithms. We achieved F1 score of 0.705 for identifying the argumentative sentences which is quite promising result and can be used as the basis for a general argument-mining framework.

A multi-component pheromone in the urine of dominant male tilapia (Oreochromis mossambicus) reduces aggression in rivals

Males often use scent to communicate their domi- nance, and to mediate aggressive and breeding behaviors. In teleost fish, however, the chemical composition of male pher- omones is poorly understood. Male Mozambique tilapia, Oreochromis mossambicus, use urine that signals social status and primes females to spawn. The urinary sex pheromone di- rected at females consists of 5β-pregnane-3α,17α,20β-triol 3- glucuronate and its 20α-epimer. The concentration of these is positively correlated with male social rank. This study tested whether dominant male urine reduces aggression in receiver males, and whether the pregnanetriol 3-glucuronates also re- duce male-male aggression. Males were allowed to fight their mirror image when exposed to either: i) water control or a chemical stimulus; ii) dominant male urine (DMU); iii) C18- solid phase (C18-SPE) DMU eluate; iv) C18-SPE DMU eluate plus filtrate; v) the two pregnanetriol 3-glucuronates (P3Gs); or vi) P3Gs plus DMU filtrate. Control males mounted an increas- ingly aggressive fight against their image over time. However, DMU significantly reduced this aggressive response. The two urinary P3Gs did not replicate the effect of whole DMU. Neither did the C18-SPE DMU eluate, containing the P3Gs, alone, nor the C18-SPE DMU filtrate to which the two P3Gs were added. Only exposure to reconstituted DMU (C18-SPE eluate plus filtrate) restored the aggression-reducing effect of whole DMU. Olfactory activity was present in the eluate and the polar filtrate in electro-olfactogram studies. We conclude that P3Gs alone have no reducing effect on aggression and that the urinary signal driving off male competition is likely to be a multi-component pheromone, with components present in both the polar and non-polar urine fractions.

Semi-implicit finite strain constitutive integration and mixed strain/stress control based on intermediate configurations

A new semi-implicit stress integration algorithm for finite strain plasticity (compatible with hyperelas- ticity) is introduced. Its most distinctive feature is the use of different parameterizations of equilibrium and reference configurations. Rotation terms (nonlinear trigonometric functions) are integrated explicitly and correspond to a change in the reference configuration. In contrast, relative Green–Lagrange strains (which are quadratic in terms of displacements) represent the equilibrium configuration implicitly. In addition, the adequacy of several objective stress rates in the semi-implicit context is studied. We para- metrize both reference and equilibrium configurations, in contrast with the so-called objective stress integration algorithms which use coinciding configurations. A single constitutive framework provides quantities needed by common discretization schemes. This is computationally convenient and robust, as all elements only need to provide pre-established quantities irrespectively of the constitutive model. In this work, mixed strain/stress control is used, as well as our smoothing algorithm for the complemen- tarity condition. Exceptional time-step robustness is achieved in elasto-plastic problems: often fewer than one-tenth of the typical number of time increments can be used with a quantifiable effect in accuracy. The proposed algorithm is general: all hyperelastic models and all classical elasto-plastic models can be employed. Plane-stress, Shell and 3D examples are used to illustrate the new algorithm. Both isotropic and anisotropic behavior is presented in elasto-plastic and hyperelastic examples.