Measuring semantic similarity is a task needed in many Natural Language Processing (NLP) applications. For example, in Machine Translation evaluation, semantic similarity is used to assess the quality of the machine translation output by measuring the degree of equivalence between a reference translation and the machine translation output. The problem of semantic similarity (Corley and Mihalcea, 2005) is de ned as measuring and recognising semantic relations between two texts. Semantic similarity covers di erent types of semantic relations, mainly bidirectional and directional. This thesis proposes new methods to address the limitations of existing work on both types of semantic relations. Recognising Textual Entailment (RTE) is a directional relation where a text T entails the hypothesis H (entailment pair) if the meaning of H can be inferred from the meaning of T (Dagan and Glickman, 2005; Dagan et al., 2013). Most of the RTE methods rely on machine learning algorithms. de Marne e et al. (2006) propose a multi-stage architecture where a rst stage determines an alignment between the T-H pairs to be followed by an entailment decision stage. A limitation of such approaches is that instead of recognising a non-entailment, an alignment that ts an optimisation criterion will be returned, but the alignment by itself is a poor predictor for iii non-entailment. We propose an RTE method following a multi-stage architecture, where both stages are based on semantic representations. Furthermore, instead of using simple similarity metrics to predict the entailment decision, we use a Markov Logic Network (MLN). The MLN is based on rich relational features extracted from the output of the predicate-argument alignment structures between T-H pairs. This MLN learns to reward pairs with similar predicates and similar arguments, and penalise pairs otherwise. The proposed methods show promising results. A source of errors was found to be the alignment step, which has low coverage. However, we show that when an alignment is found, the relational features improve the nal entailment decision. The task of Semantic Textual Similarity (STS) (Agirre et al., 2012) is de- ned as measuring the degree of bidirectional semantic equivalence between a pair of texts. The STS evaluation campaigns use datasets that consist of pairs of texts from NLP tasks such as Paraphrasing and Machine Translation evaluation. Methods for STS are commonly based on computing similarity metrics between the pair of sentences, where the similarity scores are used as features to train regression algorithms. Existing methods for STS achieve high performances over certain tasks, but poor results over others, particularly on unknown (surprise) tasks. Our solution to alleviate this unbalanced performances is to model STS in the context of Multi-task Learning using Gaussian Processes (MTL-GP) ( Alvarez et al., 2012) and state-of-the-art iv STS features ( Sari c et al., 2012). We show that the MTL-GP outperforms previous work on the same datasets.

Many technological hurdles remain before a viable commercial magnetic fusion energy reactor can be constructed, including the development of plasma-facing components with long lifetimes that can survive the harsh environment inside a reactor. One such component, the divertor, which maintains the purity of the plasma by removing fusion byproducts from the reactor, must be able to accommodate very large incident heat fluxes of at least 10 MW/m^2 during normal operation. Modular helium-cooled tungsten divertors are one of the leading divertor designs for future commercial fusion reactors, and a number of different candidates have been proposed including the modular He-cooled divertor concept with pin array (HEMP), the modular He-cooled divertor concept with multiple-jet-cooling (HEMJ), and the helium-cooled flat plate (HCFP). These three designs typically operate with helium coolant inlet temperatures of 600 °C and inlet pressures of 10 MPa. Performing experiments at these conditions to evaluate the thermal performance of each design is both challenging and expensive. An alternative, more economical approach for evaluating different designs exploits dynamic similarity. Here, geometrically similar mockups of a single divertor module are tested using coolants at lower temperatures and pressures. Dynamically similar experiments were performed on an HEMP-like divertor with helium and argon at inlet temperatures close to room temperature, inlet pressures below 1.4 MPa, and incident heat fluxes up to 2 MW/m^2. The results are used to predict the maximum heat flux that the divertor can accommodate, and the pumping power as a fraction of incident thermal power, for a given maximum tungsten temperature. A new nondimensional parameter, the thermal conductivity ratio, is introduced in the Nusselt number correlations which accounts for variations in the amount of conduction heat transfer through the walls of the divertor module. Numerical simulations of the HCFP divertor are performed to investigate how the thermal conductivity ratio affects predictions for the maximum heat flux obtained in previous studies. Finally, a helium loop is constructed and used to perform dynamically similar experiments on an HEMJ module at inlet temperatures as high as 300 °C, inlet pressures of 10 MPa, and incident heat fluxes as great as 4.9 MW/m^2. The correlations generated from this work can be used in system codes to determine optimal designs and operating conditions for a variety of fusion reactor designs.

<p>Upplevd likhet med målpersonen har ansetts vara viktig för empati och viss forskning har visat att empatin ökar med upplevd etnisk sam-hörighet. Denna studies primära syfte var att experimentellt undersöka om svenska och icke-svenska deltagare kände olika mycket empati beroende på målpersonens etnicitet samt upplevd likhet med mål-personen. Majoriteten av de 160 deltagarna rekryterades från Mälar-dalens högskola, varav 102 var svenska och 84 var kvinnliga. Resultatet visade två signifikanta disordinala interaktioner där svenska deltagare kände signifikant mer empati och upplevd likhet med en svensk än icke-svensk målperson, medan icke-svenska inte visade signifikant mer empati eller upplevd likhet med en icke-svensk än svensk målperson. Ingen signifikant skillnad i empati fanns mellan äldre och yngre deltagare. Män uppvisade signifikant lägre empati än kvinnor och inget av könen väckte mer empati. Orsaken till de disordinala interaktionerna diskuterades i termer av social kategorisering. Vidare forskning med en annan definition av begreppet etnicitet föreslogs.</p>

This paper extends recent research studying biases in venture capitalist's decision-making. We contribute to this literature by analyzing biases arising due to similarity between a venture capitalist and members of a venture team. We summarize the psychological foundations of such similarity effects and derive a set of hypotheses regarding the impact of similarity on the assessement of team quality. Using data from a conjoint experiment with 51 respondents, we find that venture capitalists tend to favor teams that are similar to themselves w.r.t. the type of training and professional experience. Our results have important implications for academics and practitioners alike. (authors' abstract)

Description Logics (DLs) are a family of logic-based knowledge representation languages used to describe the knowledge of an application domain and reason about it in formally well-defined way. They allow users to describe the important notions and classes of the knowledge domain as concepts, which formalize the necessary and sufficient conditions for individual objects to belong to that concept. A variety of different DLs exist, differing in the set of properties one can use to express concepts, the so-called concept constructors, as well as the set of axioms available to describe the relations between concepts or individuals. However, all classical DLs have in common that they can only express exact knowledge, and correspondingly only allow exact inferences. Either we can infer that some individual belongs to a concept, or we can't, there is no in-between. In practice though, knowledge is rarely exact. Many definitions have their exceptions or are vaguely formulated in the first place, and people might not only be interested in exact answers, but also in alternatives that are "close enough". This thesis is aimed at tackling how to express that something "close enough", and how to integrate this notion into the formalism of Description Logics. To this end, we will use the notion of similarity and dissimilarity measures as a way to quantify how close exactly two concepts are. We will look at how useful measures can be defined in the context of DLs, and how they can be incorporated into the formal framework in order to generalize it. In particular, we will look closer at two applications of thus measures to DLs: Relaxed instance queries will incorporate a similarity measure in order to not just give the exact answer to some query, but all answers that are reasonably similar. Prototypical definitions on the other hand use a measure of dissimilarity or distance between concepts in order to allow the definitions of and reasoning with concepts that capture not just those individuals that satisfy exactly the stated properties, but also those that are "close enough".

The investigation examined the effects of formal and semantic intralist stimulus similarity (ISS) on transfer of stimulus differentiations in the A-B, A-C paradigm.

Would you like a stranger more who shifts his/her attitudes to more closely align with yours? How would you feel if he/she aligned with everyone as opposed to just you? In Experiment 1, participants discussed with a partner disagreed upon social issues and received false feedback about whether the partner engaged in attitude alignment (shifted his/her attitude toward the participant’s attitude) following discussion. Participants also received false feedback about proportion of similarity (25%, 50%, or 75%) to the partner. Participants reported greater attraction toward partners who engaged in attitude alignment and who were more similar. However, similarity only predicted attraction in the absence of attitude alignment. Additionally, partner attitude alignment led to participant attitude alignment, and perceived reasoning ability marginally mediated the attitude alignment-attraction relationship. Similar to Experiment 1, participants in Experiment 2 received attitude alignment feedback, but they also received feedback about whether the partner engaged in attitude alignment with no others besides the participant (selective) or with many others besides the participant (unselective). Participants reported greater attraction toward partners who engaged in attitude alignment with them regardless of the partners’ attitude alignment with others. Perceived reasoning ability again mediated the attitude alignment-attraction relationship, and appeared to be more important in explaining this relation than cognitive evaluation or inferred attraction. Finally, participants reported greater trust and respect for partners who engaged in attitude alignment, but were no more willing to help those partners. This work extends our understanding of attitude alignment and its potential to affect interpersonal relationships, and it considers the influence of judgments about individuals outside of the dyad (i.e., alignment with others relative to alignment with the self).

Learning and representing semantics is one of the most important tasks that significantly contribute to some growing areas, as successful stories in the recent survey of Turney and Pantel (2010). In this thesis, we present an in- novative (and first) framework for creating a multimodal distributional semantic model from state of the art text-and image-based semantic models. We evaluate this multimodal semantic model on simulating similarity judgements, concept clustering and the newly introduced BLESS benchmark. We also propose an effective algorithm, namely Parameter Estimation, to integrate text- and image- based features in order to have a robust multimodal system. By experiments, we show that our technique is very promising. Across all experiments, our best multimodal model claims the first position. By relatively comparing with other text-based models, we are justified to affirm that our model can stay in the top line with other state of the art models. We explore various types of visual features including SIFT and other color SIFT channels in order to have prelim- inary insights about how computer-vision techniques should be applied in the natural language processing domain. Importantly, in this thesis, we show evi- dences that adding visual features (as the perceptual information coming from...

Estudos empíricos [1],[2] demonstraram que o trafego das redes Internet Protocol (IP) possui propriedades fractais tais como impulsividade, auto-similaridade e dependência de longa duração em diversas escalas de agregação temporal, na faixa de milissegundos a minutos. Essas características tem motivado o desenvolvimento de novos modelos fractais de teletráfego e de novos algoritmos de controle de trafego em redes convergentes. Este trabalho propõe um novo modelo de trafego no espaço de estados baseado numa aproximação finito-dimensional do processo AutoRegressive Fractionally Integrated Moving Average (ARFIMA). A modelagem por meio de processos auto-regressivos (AR) também é investigada. A analise estatística de series simuladas e de series reais de trafego mostra que a aplicação de modelos AR de ordem alta em esquemas de previsão de teletráfego é fortemente prejudicada pelo problema da identificação da ordem do modelo. Também demonstra-se que a modelagem da memória longa pode ser obtida as custas do posicionamento de um ou mais pólos nas proximidades do circulo de raio unitário. Portanto, a implementação do modelo AR ajustado pode ser instável devido a efeitos de quantização dos coeficientes do filtro digital. O modelo de memória longa proposto oferece as seguintes vantagens: a) possibilidade de implementação pratica, pois não requer memória infinita, b) modelagem (explícita) da região das baixas freqüências do espectro e c) viabilização da utilização do filtro de Kalman. O estudo de caso apresentado demonstra que é possível aplicar o modelo de memória longa proposto em trechos estacionários de sinais de teletráfego fractal. Os resultados obtidos mostram que a dinâmica do parâmetro de Hurst de sinais de teletráfego pode ser bastante lenta na pratica. Sendo assim, o novo modelo proposto é adequado para esquemas de previsão de trafego, tais como Controle de Admissão de Conexões (CAC) e alocação dinâmica de banda, dado que o parâmetro de Hurst pode ser estimado em tempo real por meio da aplicação da transformada wavelet discreta (Discrete Wavelet Transform (DWT)). / Empirical studies [1],[2] demonstrated that heterogeneous IP traffic has fractal properties such as impulsiveness, self-similarity, and long-range dependence over several time scales, from miliseconds to minutes. These features have motivated the development of new traffic models and traffic control algorithms. This work presents a new state-space model for teletraffic which is based on a finite-dimensional representation of the ARFIMA random process. The modeling via AutoRegressive (AR) processes is also investigated. The statistical analysis of simulated time series and real traffic traces show that the application of high-order AR models in schemes of teletraffic prediction can be highly impaired by the model identification problem. It is also demonstrated that the modeling of the long memory can be obtained at the cost of positioning one or more poles near the unit circle. Therefore, the implementation of the adjusted AR model can be unstable due to the quantization of the digital filter coefficients. The proposed long memory model has the following advantages: a) possibility of practical implementation, inasmuch it does not require infinite memory, b) explicit modeling of the low frequency region of the power spectrum, and c) forecasts can be performed via the Kalman predictor. The presented case study suggests one can apply the proposed model in periods where stationarity can be safely assumed. The results indicate that the dynamics of the Hurst parameter can be very slow in practice. Hence, the new proposed model is suitable for teletraffic prediction schemes, such as CAC and dynamic bandwidth allocation, given that the Hurst parameter can be estimated on-line via DWT.

Lança-se aqui a hipótese de que é possível reproduzir o comportamento do protótipo a partir de modelos feitos de materiais diferentes dos utilizados na estrutura de referência. A influência de distorção de espessura, de diferentes densidades, tensões de escoamento, encruamento e viscoplasticidade são todas avaliadas através de equações analíticas e simulações das estruturas básicas como viga, placa e o impacto de uma estrutura de chapa dupla. Adicionalmente, são realizadas caracterizações quase-estáticas e dinâmicas de material e experimentos em placas circulares para três escalas diferentes (1/1, 2/3 e 1/3), quatro materiais (alumínio, titânio, aço inox e cobre) e duas condições de carregamento: impacto de uma massa a baixa velocidade (3,5m/s) e um projétil a alta velocidade (130m/s). Mostra-se que o uso da lei de Johnson-Cook para descrever o comportamento viscoplástico dos materiais permite obter boa aproximação para o cálculo da velocidade de impacto corrigida, sem que sejam necessárias informações adicionais do comportamento da estrutura. Além disso, os resultados revelam a importância de se manter a razão entre massa de impacto e massa da estrutura constante em modelo e protótipo, levando a desejável similaridade das distorções entre ambas. Assim, ao se confirmar a hipótese deste trabalho, mostra-se possível inferir o comportamento de um protótipo com uso de modelos em escala reduzida feitos a partir de materiais diferentes. / It is hypothesized that it is possible to reproduce the behaviour of the prototype by using models made of different materials. The influence of thickness, different densities, flow stresses, hardening and viscoplasticity are all evaluated through analytical equations and simulations of basic structures such as beam, plate and the impact of a double plate structure. In addition, quasi-static and dynamic materials characterization are conducted , with experiments being performed in circular plates for three different scales (1/1, 2/3 and 1/3), four materials (Aluminum, Titanium, Stainless Steel and Copper) and two conditions impact of a mass (3,5m/s) and a projectile at high speed (130m/s). It is shown that the use of Johnson-Cook\'s law to describe the viscoplastic behavior of the materials allows a satisfactory approximation to the calculation of the corrected impact velocity, without the need for additional information on the behavior of the structure. In addition, the results reveal the relevance of maintaining the ratio between impact mass and structure mass constant in model and prototype, leading to the desirable similarity of the distortions between the two. Thus, when confirming the hypothesis of this work, it is possible to infer the behavior of a prototype using small scale models made from different materials.