published_or_final_version / Ecology and Biodiversity / Doctoral / Doctor of Philosophy

The paper describes an approach to improving classification-based subject access in a library catalogue. In order to enhance the use of UDC numbers in information retrieval, the authors have represented classification with thesaurus descriptors and implemented this solution in an automated way. In addition, descriptors in more than one language were used to interface classification. The authors present a solution implemented in a BiblioPhil library system. The standard formats used are UNIMARC for bibliographic and subject authority records (i.e. the UDC-based multilingual thesaurus) with MARCXML support for data transfer. The multilingual thesaurus was built according to the existing standards, the constituent parts of the classification notations being used as the basis for search terms in the multilingual information retrieval. The verbal equivalents, descriptors and non-descriptors, are used to expand the number of concepts and are given in Romanian, English and French. The authors illustrate how this approach saves the time of the indexer and provides more user-friendly and easier access to the bibliographic information. The multilingual aspect of the thesaurus enhances information access for a greater number of online users.

No description available.

No description available.

No description available.

No description available.

No description available.

Security defects or vulnerabilities are inescapable in software development. Thus, it is always better to address security issues during the software development phases, rather than developing patches after the security threats are already in place. In line with this, a number of secure software development approaches have been proposed so far to address the security issues during the development processes. However, most of these approaches lack specific process improvement activities. The practice of taking adequate corrective measures at the earliest possible time by learning from the past mistakes is absent in case of such security-aware iterative software development processes. As one might imagine, software security defect data provide an invaluable source of information for a software development team. This thesis aims at investigating existing security defect classification schemes and providing a structured security-specific defect classification and analysis methodology. Our methodology which we build on top of the Orthogonal Defect Classification (ODC) scheme, is customized to generate in-process feedback by analyzing security defect data. More specifically, we perform a detailed analysis on the classified security defect data and obtain in-process feedback using which the next version of software can be more secure and reliable. We experiment our methodology on the Mozilla Firefox and Chrome security defect repositories using six consecutive versions and milestones, respectively. We find that the in-process feedback generated by applying this methodology can help take corrective actions as early as possible in iterative secure software development processes. Finally, we study the correlations between software security defect types and the phases of software development life-cycle to understand development improvement by complementing the previous ODC scheme. / Thesis (Master, Computing) -- Queen's University, 2012-10-30 15:47:34.47

In this thesis, we focus on the problem of image classification and investigate how its performance can be systematically improved. Improving the performance of different computer vision methods has been the subject of many studies. While different studies take different approaches to achieve this improvement, in this thesis we address this problem by investigating the relevance of the statistics collected from the image. We propose a framework for gradually improving the quality of an already existing image descriptor. In our studies, we employ a descriptor which is composed the response of a series of discriminative components for summarizing each image. As we will show, this descriptor has an ideal form in which all categories become linearly separable. While, reaching this form is not possible, we will argue how by replacing a small fraction of these components, it is possible to obtain a descriptor which is, on average, closer to this ideal form. To do so, we initially identify which components do not contribute to the quality of the descriptor and replace them with more robust components. As we will show, this replacement has a positive effect on the quality of the descriptor. While there are many ways of obtaining more robust components, we introduce a joint feature selection problem to obtain image features that retains class discriminative properties while simultaneously generalising between within class variations. Our approach is based on the concept of a joint feature where several small features are combined in a spatial structure. The proposed framework automatically learns the structure of the joint constellations in a class dependent manner improving the generalisation and discrimination capabilities of the local descriptor while still retaining a low-dimensional representations. The joint feature selection problem discussed in this thesis belongs to a specific class of latent variable models that assumes each labeled sample is associated with a set of different features, with no prior knowledge of which feature is the most relevant feature to be used. Deformable-Part Models (DPM) can be seen as good examples of such models. These models are usually considered to be expensive to train and very sensitive to the initialization. Here, we focus on the learning of such models by introducing a topological framework and show how it is possible to both reduce the learning complexity and produce more robust decision boundaries. We will also argue how our framework can be used for producing robust decision boundaries without exploiting the dataset bias or relying on accurate annotations. To examine the hypothesis of this thesis, we evaluate different parts of our framework on several challenging datasets and demonstrate how our framework is capable of gradually improving the performance of image classification by collecting more robust statistics from the image and improving the quality of the descriptor. / <p>QC 20140506</p>

The phylogeny of the Ammobatini was studied, with regard to the principles of cladistics using parsimony, and the classification is revised. It is concluded that the tribe forms a monophyletic group that comprises six distinct monophyletic genera: Pasite Jurine, Sphecodopsis Bischoff, Ammobates Latreille, Melanempis Saussure, Spinopasites Warncke and Oreopasites Cockerell, of which Pasites, Sphecodopsis, Ammobates and Melanempis occur in the Afrotropical Region. The Afrotropical species of these four genera are revised. Pseudopasites Bischoff and Pseudodichroa Bischoff are synonymized with Sphecodopsis. Pasites includes 17 Afrotropical species, Sphecodopsis 10 species, and Ammobates and Melanempis are each known from a single Afrotropical species. Ten new species are described: Passites nilssoni, P. paulyi, P. humecta, P. gnoma, P. namibiensis, P. somalica, Sphecodopsis vespericena, S. longipygidium, S. namaquensis and Ammobates auster. Thirty-three names are synonymized: they are P. nigerrima (Friese), P. argentata (Baker) (= P. barkeri (Cockerell)); P. chubbi Cockerell, P. nigritula Bischoff, P. peratra Cockerell (= P. atra Friese); P. nigripes (Friese), P. fortis Cockerell, P. subfortis Cockerell, P. stordyi Cockerell, P. voiensis Cockerell, P. altior Cockerell (= P. carnifex (Gerstaecker)); P. natalensis (Cockerell), P. aiboguttatus (Friese), P. ogilviei (Cockerell) (= P. jenseni (Friese)); P. alivalensis (Cockerell), P. rufitarsis (Cockerell) (= P. histrio (Gerstaecker)); P. marshaUi (Cockerell) (= P. jonesi (Cockerell)); P. abessinica (Friese), P. fulviventris (Bischoff), P. rhodesialla (Bischoff), P. apicalis (Bischoff), P. turneri (Cockerell), P.politula (Cockerell), P. indecisa (Cockerell), P. nudicauda (Cockerell), P. bechuanica (Cockerell), P. breviceps (Cockerell) (= P. appletoni (Cockerell); S. rufula (Cockerell) (= S. minutissima (Cockerell)); S. pygmaea (Friese), S. rufescens Bischoff, S. algoensis Bischoff, S. perpunctata Cockerell (= S. aculeata (Friese)); S. leonis (Cockerell) (= S. semirufa (Cockerell)). Keys to the genera and species are provided. / Thesis (Ph.D.)-University of Natal, Pietermaritzburg, 1994.