Anyone can upload news instantaneously to Twitter in 140-characters or less, therefore it is important to assess the credibility of tweets, particularly during a foodborne illness outbreak. According to a Microsoft study there are numerous Twitter features that impact the credibility of tweets. This study examines students’ use of Twitter, and their perceptions of how features and sources impact the credibility of tweets related to a foodborne illness. An online survey was completed by Texas A&M University students classified as U3 juniors (N = 200) in social science-based majors in the College of Agriculture and Life Sciences. The study revealed students to be moderate users of Twitter. Eight features were identified to impact credibility: verified author topic expertise, account has verification seal, tweet contains grammar/punctuation mistakes, author is someone you’ve heard of, author is often mentioned/retweeted, author often tweets on topic, and author has many followers. Three Twitter identities a professor, student, and student organization, were created to measure tweet credibility. Tweets from the professor were perceived more credible than tweets attributed to student or student organization. The results indicated statistical differences between features attributed to each source. These findings can help sources determine what features can make tweets the most credible. This study has important implications for organizations that engage consumers on Twitter when breaking news such as a foodborne illness outbreak occurs.

In this thesis we consider the task of catching a moving target with multiple pursuers, also known as the “Pursuit Game”, in which coordination among the pursuers is critical. Our testbed is inspired by the pursuit problem in video games, which require fast planning to guarantee fluid frame rates. We apply supervised machine learning methods to automatically derive efficient multi-agent pursuit strategies on rectangular grids. Learning is achieved by computing training data off-line and exploring the game tree on small problems. We also generalize the data to previously unseen and larger problems by learning robust pursuit policies, and run empirical comparisons between several sets of state features using a simple learning architecture. The empirical results show that 1) the application of learning across different maps can help improve game-play performance, especially on non-trivial maps against intelligent targets, and 2) simple heuristic works effectively on simple maps or less intelligent targets.

Although the positional accuracy of spatial data has long been of fundamental importance in GIS, it is still largely unknown for linear features. This is compromising the ability of GIS practitioners to undertake accurate geographic analysis and hindering GIS in fulfilling its potential as a credible and reliable tool. As early as 1987 the US National Center for Geographic Information and Analysis identified accuracy as one of the key elements of successful GIS implementation. Yet two decades later, while there is a large body of geodetic literature addressing the positional accuracy of point features, there is little research addressing the positional accuracy of linear features, and still no accepted accuracy model for linear features. It has not helped that national map and data accuracy standards continue to define accuracy only in terms of “well-defined points”. This research aims to address these shortcomings by exploring the effect on linear feature positional accuracy of feature type, complexity, segment length, vertex proximity and e-scale, that is, the scale of the paper map from which the data were originally captured or to which they are customised for output. / The research begins with a review of the development of map and data accuracy standards, and a review of existing research into the positional accuracy of linear features. A geographically sensible error model for linear features using point matching is then developed and a case study undertaken. Features of five types, at five e-scales, are selected from commonly used, well-regarded Australian topographic datasets, and tailored for use in the case study. Wavelet techniques are used to classify the case study features into sections based on their complexity. Then, using the error model, half a million offsets and summary statistics are generated that shed light on the relationships between positional accuracy and e-scale, feature type, complexity, segment length, and vertex proximity. Finally, auto-regressive time series modelling and moving block bootstrap analysis are used to correct the summary statistics for correlation. / The main findings are as follows. First, metadata for the tested datasets significantly underestimates the positional accuracy of the data. Second, positional accuracy varies with e-scale but not, as might be expected, in a linear fashion. Third, positional accuracy varies with feature type, but not as the rules of generalisation suggest. Fourth, complex features lose accuracy faster than less complex features as e-scale is reduced. Fifth, the more complex a real-world feature, the worse its positional accuracy when mapped. Finally, accuracy mid-segment is greater than accuracy end-segment.

Despite recent advances, improving the accuracy of automatic speaker recognition systems remains an important and challenging area of research. This thesis investigates two-phase based features, namely the frequency modulation (FM) feature and the group delay feature in order to improve the speaker recognition accuracy. Introducing features complementary to spectral envelope-based features is a promising approach for increasing the information content of the speaker recognition system. Although phase-based features are motivated by psychophysics and speech production considerations, they have rarely been incorporated into speaker recognition front-ends. A theory has been developed and reported in this thesis, to show that the FM component can be extracted using second-order all pole modelling, and a technique for extracting FM features using this model is proposed, to produce very smooth, slowly varying FM features that are effective for speaker recognition tasks. This approach is shown herein to significantly improve speaker recognition performance over other existing FM extraction methods. A highly computationally efficient FM estimation technique is then proposed and its computational efficiency is shown through a comparative study with other methods with respect to the trade off between computational complexity and performance. In order to further enhance the FM based front-end specifically for speaker recognition, optimum frequency band allocation is studied in terms of the number of sub-bands and spacing of centre frequencies, and two new frequency band re-allocations are proposed for FM based speaker recognition. Two group delay features are also proposed: log compressed group delay feature and the sub-band group delay feature, to address problems in group delay caused by the zeros of the z-transform polynomial of a speech signal being close to the unit circle. It has been shown that the combination of group delay and FM, complements Mel Frequency Cepstral Coefficient (MFCC) in speaker recognition tasks. Furthermore, the proposed FM feature is successfully utilised for automatic forensic speaker recognition, which is implemented based on the likelihood ratio framework with two stage modelling and calibration, and shown to behave in a complementary manner to MFCCs. Notably, the FM based system provides better calibration loss than the MFCC based system, suggesting less ambiguity of FM information than MFCC information in an automatic forensic speaker recognition system. In order to demonstrate the effectiveness of FM features in a large scale speaker recognition environment, an FM-based speaker recognition subsystem is developed and submitted to the NIST 2008 speaker recognition evaluation as part of the I4U submission. Post evaluation analysis shows a 19.7% relative improvement over the traditional MFCC based subsystem when it is augmented by the FM based subsystem. Consistent improvements in performance are obtained when MFCC is augmented with FM in all sub-categories of NIST 2008, in three development tasks and for the NIST 2001 database, demonstrating the complementary behaviour of MFCC and FM features.

Thesis (Ph. D.)--University of Texas at Austin, 2007. / Vita. Includes bibliographical references.

Thesis (Ph. D.)--Ohio State University, 2004. / Document formatted into pages; contains xxi, 371 p. Includes bibliographical references (p. 324-371). Abstract available online via OhioLINK's ETD Center; full text release delayed at author's request until 2007 Aug. 18.

Thesis (Ph. D.)--University of Washington, 1996. / Vita. Includes bibliographical references (leaves [206]-219).

abstract: When consumers make experiential purchases, they often have to decide between experiences that contain many or few features. Contrary to prior research demonstrating that consumers prefer feature-rich products before consumption but feature-poor products after consumption, the author reveals a reversal of this effect for experiences. Specifically, the author hypothesizes and finds that consumers prefer feature-poor experiences before consumption (a phenomenon denoted as `feature apprehension') but prefer feature-rich experiences after consumption. This feature apprehension occurs before consumption because consumers are concerned with the uncertainty associated with attaining a satisfying outcome from the experience. Manipulating the temporal distance with which consumers view the experience can attenuate this effect. Additionally, locus of control and social signaling moderate consumers' post-consumption preference for feature-rich experiences. The author proposes several recommendations for consumers and providers of experiences. / Dissertation/Thesis / Doctoral Dissertation Business Administration 2015

O polietileno linear de baixa densidade (LLDPE) com comonômeros 1-buteno e 1-octeno apresenta fácil processabilidade e baixo custo. Por outro lado, as poliolefinas em geral possuem baixa reatividade, dificultando a adesão com outros substratos e limitando suas aplicações. Por esta razão tratamentos foram desenvolvidos para minimizar essa característica, sendo conhecidos como processos de enxertia ou funcionalização. Este trabalho teve por finalidade avaliar a influência do tamanho dos comonômeros buteno e octeno e dos dois tipos de agentes funcionalizantes anidrido maléico (AM) e anidrido tetrahidroftálico (ATF) no processo de funcionalização via úmida das poliolefinas. Para se avaliar a significância estatística de quatro variáveis dos experimentos: porcentagem do agente funcionalizante (AM ou ATF), porcentagem do iniciador peróxido de benzoíla (BPO), tipo de comonômero (buteno e octeno) e o tempo de reação, usou-se as ferramentas estatísticas conhecidas como planejamento fatorial fracionário e planejamento fatorial composto central. O grau de funcionalização das poliolefinas modificadas foi determinado por espectroscopia vibracional no infravermelho (IR). Os efeitos do grau de funcionalização e do tipo de anidrido sobre o grau de cristalinidade das amostras foram obtidos por difração de raios-X e calorimetria exploratória diferencial (DSC). A molhabilidade das poliolefinas foi determinada, antes e depois da funcionalização, por medidas de ângulo de contato, utilizando-se gotas de água como líquido. Observamos que (1) a funcionalização das poliolefinas modificadas com AM se mostrou mais eficiente do que com ATF, porque as moléculas de ATF são mais reativas, o que as torna mais suscetíveis a reações paralelas como a homopolimerização e/ou reticulação; (2) a modificação das poliolefinas com AM apresentou significância estatística só nas variáveis: porcentagem de BPO e porcentagem de AM; (3) as, poliolefinas modificadas com ATF apresentaram duas variáveis significativas, o tempo da reação e a interação existente entre o tempo de reação e a porcentagem de ATF; (4) as poliolefinas modificadas com AM e ATF apresentaram uma redução na temperatura de fusão de até 3,3°C e 4,3°C respectivamente, evidenciando alterações na estrutura cristalina do polímero; (5) ensaios de difração de raios-X corroboram com os valores obtidos por calorimetria exploratória diferencial (DSC), uma vez que não ocorreu deslocamento dos máximos de difração e tampouco do halo amorfo; mas houve uma diminuição acentuada nos picos de difração (diminuição da altura e aumento da largura), reflexo da desestruturação causada pelo agente funcionalizante nos domínios cristalinos; (6) observou-se que o aumento no GF promoveu uma maior molhabilidade da superfície, indicando uma maior interação dos grupos polares enxertados com a água. Os valores altos de Li8 indicam alta rugosidade e ou heterogeneidade química da superfície; (7) não houve variação significativa no grau de funcionalização quando utilizado poliolefinas com diferentes comonômeros; (8) o aumento na concentração de poliolefina promove uma maior funcionalização. (9) o método de funcionalização desenvolvido levou à reticulação mínima da poliolefinas. / Linear low-density polyethylene with comonomer 1-butene and 1-octene presents the advantages of easy processing and low cost. On the other hand, polyolefins are inert materials with very low surface energy, impeaching the adhesion to other materials. In order to improve the adhesion to other materials, grafting or functionalization processes have been developed. This work evaluates the influence of the comonomers butene and octene and of two types of anhydride, maleic anhydride (MA) and tetrahydrophthalic anhydride (THA), on the functionalization of polyolefins by means of chemical reactions in solution. In order to evaluate the statistic meaning of four variables (content of anhydride, content of benzoyl peroxide, type of comonomer, and time of reaction) in the functionalization process the fractional factorial design and the central composite experimental design were used. The degree of functionalization (DF) was determined by means of infrared spectroscopy (IR). The effects of (DF) and of the type of anhydride on the crystallinity of the modified samples were obtained from X-ray diffraction and differential scanning calorimetry (DSC). The wettability of polyolefins was determinated before and after of the functionalization by means of contact angle measurements using water drops like liquid. We observe that ( 1) the functionalization of polyolefins modified with MA are more efficient than the polyolefins modified with THA because the THA molecules are more reactive than the MA molecules, favoring side reactions as homopolymerization and cross-linking; (2) the modification of the polyolefins with MA shows statistic meaning only in the following variables: contents of BPO and MA percentage; (3) the polyolefins modified with THA show significative variables: the time of reaction and the interaction between the time of reaction and the THA content; (4) the polyolefins modified with MA and THA show a reduction in the melt temperature of 3,3°C and 4,3°C respectively, indicating changes in the crystalline structure of polymer; (5) X-ray diffractograms confirmed the results obtained by differential scanning calorimetry (DSC), since the diffraction peaks turned smaller; (6) the increase in the DF improved the surface wettability; (7) the size of comonomer has no influence on the DF; (8) the DF depends on the polyolefin concentration; (9) the polyolefins modification by chemical reactions in solution presented very low cross-linking degree.

Life-history features of terrestrial mammals in the southern African subregion were explored in an attempt to identify the factor(s) that may influence the evolution of common mammalian life-history patterns within this subregion. Allometry and phylogeny were found to underly most of the variation observed in life-history features. Residual variation was considered and age at maturity, gestation period and litter size emerged as additional features influencing life-history patterns. Southern African mammals could generally be arranged along a theoretical r-K continuum with small mammals having relatively short developmental periods and relatively large litter sizes represented at one extreme and large mammals having relatively long developmental periods with relatively small litter sizes at the opposite extreme. However, the carnivores did not conform to this expected general gradient suggesting that they may be an exceptional order as far as their life-histories are concerned. / Dissertation (MSc)--University of Pretoria, 2011. / Zoology and Entomology / unrestricted