This thesis is concerned with improving the effectiveness of nearest neighbour search. Nearest neighbour search is the problem of finding the most similar data-points to a query in a database, and is a fundamental operation that has found wide applicability in many fields. In this thesis the focus is placed on hashing-based approximate nearest neighbour search methods that generate similar binary hashcodes for similar data-points. These hashcodes can be used as the indices into the buckets of hashtables for fast search. This work explores how the quality of search can be improved by learning task specific binary hashcodes. The generation of a binary hashcode comprises two main steps carried out sequentially: projection of the image feature vector onto the normal vectors of a set of hyperplanes partitioning the input feature space followed by a quantisation operation that uses a single threshold to binarise the resulting projections to obtain the hashcodes. The degree to which these operations preserve the relative distances between the datapoints in the input feature space has a direct influence on the effectiveness of using the resulting hashcodes for nearest neighbour search. In this thesis I argue that the retrieval effectiveness of existing hashing-based nearest neighbour search methods can be increased by learning the thresholds and hyperplanes based on the distribution of the input data. The first contribution is a model for learning multiple quantisation thresholds. I demonstrate that the best threshold positioning is projection specific and introduce a novel clustering algorithm for threshold optimisation. The second contribution extends this algorithm by learning the optimal allocation of quantisation thresholds per hyperplane. In doing so I argue that some hyperplanes are naturally more effective than others at capturing the distribution of the data and should therefore attract a greater allocation of quantisation thresholds. The third contribution focuses on the complementary problem of learning the hashing hyperplanes. I introduce a multi-step iterative model that, in the first step, regularises the hashcodes over a data-point adjacency graph, which encourages similar data-points to be assigned similar hashcodes. In the second step, binary classifiers are learnt to separate opposing bits with maximum margin. This algorithm is extended to learn hyperplanes that can generate similar hashcodes for similar data-points in two different feature spaces (e.g. text and images). Individually the performance of these algorithms is often superior to competitive baselines. I unify my contributions by demonstrating that learning hyperplanes and thresholds as part of the same model can yield an additive increase in retrieval effectiveness.

The potential for conjugated linoleic acid (CLA) incorporation into pressure-sensitive adhesive (PSA) formulations was evaluated. A series of free radical bulk copolymerizations of CLA/styrene (Sty) and CLA/butyl acrylate (BA) were designed to allow the estimation of reactivity ratios. Bulk terpolymerizations of CLA/Sty/BA were also evaluated before moving to emulsion terpolymerizations of CLA/Sty/BA. The polymers were characterized for composition, conversion, molecular weight and glass transition temperature while latexes were characterized for viscosity, particle size, tack, peel strength, and shear strength. All experiments were performed at 80oC and monitored with attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy. While bulk experiments were monitored off-line, the emulsion experiments were monitored in-line. Absorbance peaks related to the monomers and polymer were tracked to provide conversion and polymer composition data using a multivariate calibration method. Off-line measurements using gravimetry and 1H-NMR spectroscopy were compared to the ATR-FTIR data and no significant differences were detected between the measurement methods. Pseudo-kinetic models, developed and validated with the copolymer experimental data, were used to estimate reactivity ratios. The copolymer pseudo-kinetic models were extended to a terpolymer pseudo-kinetic model and validated with experimental data. The pseudo-kinetic models incorporated the ability of oleic acid, a common impurity found in CLA, to trap electrons thus influencing the reaction kinetics significantly. The influence of terpolymer composition, chain transfer agent concentration, cross-linker concentration, molecular weight, viscosity and particle size on tack, peel strength and shear strength was investigated by using a constrained mixture design. The final forms of the resulting empirical models allowed the creation of 3D response surfaces for PSA performance optimization. The incorporation of 30 wt.% CLA into a practical PSA application suitable for the removable adhesives category was achieved.

Cross-site scripting (XSS) flaws are a class of security flaws particular to web applications. XSS flaws generally allow an attacker to supply affected web application with a malicious input that is then included in an output page without being properly encoded (sanitized). Recent advances in web applica- tion technologies and web browsers introduced various prevention mechanisms, narrowing down the scope of possible XSS attacks, but those mechanisms are usually selective and prevent only a subset of XSS flaws. Among the types of XSS flaws that are largely omitted are the context- sensitive XSS flaws. A context-sensitive XSS flaw occurs when the potentially malicious input is sanitized by the affected web application before being included in the output page, but the sanitization is not appropriate for the browser con- text of the sanitized value. Another type of XSS flaws, which is already better known, but still insufficiently prevented, are the stored XSS flaws. Applica- tions affected by the stored XSS flaws store the unsafe client input in persistent storage and return it in another HTTP response to (possibly) another client. Our work is focused on advancing state-of-the-art automated detection of those two types of XSS flaws using various analysis techniques ranging from purely static analysis to dynamic graybox analysis.

Liposomal doxorubicin is a clinically important drug formulation indicated for the treatment of several different forms of cancer. For doxorubicin to exert a therapeutic effect, it must gain access to the nucleus. However, a large proportion of the liposomal doxorubicin dose fails to work because it is sequestered within endolysosomal organelles following endocytosis of the liposomes due to the phenomenon of ion trapping. Listeriolysin O (LLO) is a pore-forming protein that can provide a mechanism for endosomal escape. The present study demonstrates that liposomal coencapsulation of doxorubicin with LLO enables a significantly larger percentage of the dose to colocalize with the nucleus compared to liposomes containing doxorubicin alone. The change in intracellular distribution resulted in a significantly more potent formulation of liposomal doxorubicin as demonstrated in both the ovarian carcinoma cell line A2780 and its doxorubicin-resistant derivative A2780ADR.

Cette recherche a eu comme objectif de comprendre le concept de «sensible» dans la sémiotique greimasienne et post-greimasienne, sous la perspective de l'historiographie linguistique, en contextualisant son apparition et sa permanence dans les études sémiotiques contemporaines. Dans ce travail, le «sensible» est défini comme un hyperonyme et les autres concepts circonscrits dans son champ sont considérés comme ses domaines : de la corporéité, de la passionnalité et de la sensibilité. Pour chaque domaine, nous avons souligné les termes qui concernent le sensible : corps, affection, passion, émotion, contagion, sensation, perception, esthétique et estésique. On a récupéré le contexte théorique de ces concepts par les principes historiographiques de la contextualisation, l’immanence, l'adéquation et de l'influence, de K. Koerner (1996, 2014), les paramètres de couverture, la perspective et la profondeur et les types de composants heuristiques, herméneutique et de reconstruction systématique de P. Swiggers (2009, 2015) et l’horizon de rétrospection de S. Auroux (2008), en traçant son chemin depuis ses origines en Sémantique Structurale (1966), de A. J. Greimas, et par l'émergence et son impact dans les travaux de J. Fontanille, E. Landowski et C. Zilberberg, qui correspondent à la période que nous avons appelé ici post-greimassienne. Ensuite, on a défini dans quelle mesure le «sensible» a apparu dans la rhétorique et / ou l'immanence des œuvres de sémioticiens choisis. Après avoir établi les déploiements épistémologiques du «sensible», finalement, nous avons défini le lieu historique et épistémologique d'une sémiotique, considérée aujourd’hui sensible ou plus sensible, en explicitant sa pertinence dans les études du langage. / This research aimed to understandthe concept of “sensitive”in greimasian and post-greimasian semiotics, due to the bias of linguistic historiography, contextualizing its emergence and permanence in contemporary semiotic studies. In this work, the “sensitive”is defined as a hyperonym and the other concepts circumscribed in its field are seenas its domains: of corporality, of passion and of sensibility. In each domain, we highlight terms related to the sensitive: body, affection, passion, emotion, contagion, sensation, perception, esthetics and aesthetics. We retrieve the theoretical thickness of these concepts through the bias of the historiographic principles of contextualization, immanence, adequacy and influence, by K. Koerner (1996, 2014a), the coverage parameters, perspective and depth, heuristic, hermeneutic and reconstruction-systematic component typesP. Swiggers (2009, 2015), SO Murray specialties groups(1994, 1998) and retrospection horizonsS. Auroux (2008), tracing his course since his origins in Structural Semantics (1966) by A. J.Greimas, and passing through the emergency and its repercussion in the works of J. Fontanille, E. Landowski and C. Zilberberg, which correspond to the period we call post-Greimasian. Then we define to what extent the “sensitive”appeared in the rhetoric and / or immanence of the works of the chosen semioticists. After establishing the epistemological unfolding of the “sensitive”, finally, we were able to define the historical and epistemological place of a semiotics, considered today as sensitive or more sensitive, explaining its relevance in language studies

No description available.

<p> Temperature-sensitive mutants of the bacteriophage PBS 2 were isolated from lysates treated with various mutagens. Complementation tests assigned the mutants to 10 cistrons. The mutants were mapped by two factor crosses and formed a linear map approximately 50 recombination percent in length. </p> <p> A method for phage transformation was developed. By the use of wild type DNA fragments fractionated according to their guanine + cytosine content in HgCs₂SO₄ gradients, it was possible to determine the base composition of certain regions of the chromosomes. </p> / Thesis / Doctor of Philosophy (PhD)

<p> Temperature sensitive microgel latexes of poly (N-isopropylacrylamide) cross-linked with N-N'methylene bisacrylamide (BA) were prepared and characterized by Dynamic Light Scattering, Titration and Electrophoresis. The study of gels, including temperature sensitive gels, is limited by the large size of traditional bulk gels which are slow to respond to changes and are difficult to measure. An alternative system, which may be easier to study, is a microgel latex which would constitute small particles of gel which would respond quickly to changes in their environment and could also be measured using colloidal measuring techniques like dynamic light scattering and particle electrophoresis.</p> <p> Monodisperse and stable microgel latex particles were prepared by reacting Nisopropylacrylamide (NIPAM) monomer with a cross-linking agent BA in water at 70°C with a surfactant (sodium dodecylsulfate) present. Latexes prepared without surfactant were polydisperse.</p> <p> Characterization of the poly (NIPAM) particles by dynamic light scattering at several different temperatures showed that the particles go through a transition from a water swollen gel at low temperature to a shrunken gel with a low water content at high temperature. The transition occurs about 32°C. The degree of swelling of the poly (NIPAM) particles can be expressed by the Flory-Huggins Interaction parameter c and is dependent upon the level of cross-linking agent included.</p> <p> Titration and electrophoresis results indicate that the particles contain about 0.39 Coulombs per gram of polymer of carboxylic and sulfuric charged end groups which are distributed throughout the particle.</p> / Thesis / Master of Engineering (MEngr)

No description available.

No description available.