Support Vector Machines have been used to do classification and regression analysis. One important part of SVMs are the kernels. Although there are several widely used kernel functions, a carefully designed kernel will help to improve the accuracy of SVMs. We present two methods in terms of customizing kernels: one is combining existed kernels as new kernels, the other one is to do feature selection. We did theoretical analysis in the interpretation of feature spaces of combined kernels. Further an experiment on a chemical data set showed improvements of a linear-Gaussian combined kernel over single kernels. Though the improvements are not universal, we present a new idea of creating kernels in SVMs.

Support Vector Machines have been used to do classification and regression analysis. One important part of SVMs are the kernels. Although there are several widely used kernel functions, a carefully designed kernel will help to improve the accuracy of SVMs. We present two methods in terms of customizing kernels: one is combining existed kernels as new kernels, the other one is to do feature selection. We did theoretical analysis in the interpretation of feature spaces of combined kernels. Further an experiment on a chemical data set showed improvements of a linear-Gaussian combined kernel over single kernels. Though the improvements are not universal, we present a new idea of creating kernels in SVMs.

No description available.

In recent years, machine learning has emerged as an important discipline. However, despite the popularity of machine learning techniques, data in the form of discrete structures are not fully exploited. For example, when data appear as graphs, the common choice is the transformation of such structures into feature vectors. This procedure, though convenient, does not always effectively capture topological relationships inherent to the data; therefore, the power of the learning process may be insufficient. In this context, the use of kernel functions for graphs arises as an attractive way to deal with such structured objects. On the other hand, several entities in computational biology applications, such as gene products or proteins, may be naturally represented by graphs. Hence, the demanding need for algorithms that can deal with structured data poses the question of whether the use of kernels for graphs can outperform existing methods to solve specific computational biology problems. In this dissertation, we address the challenges involved in solving two specific problems in computational biology, in which the data are represented by graphs. First, we propose a novel approach for protein function prediction by modeling proteins as graphs. For each of the vertices in a protein graph, we propose the calculation of evolutionary profiles, which are derived from multiple sequence alignments from the amino acid residues within each vertex. We then use a shortest path graph kernel in conjunction with a support vector machine to predict protein function. We evaluate our approach under two instances of protein function prediction, namely, the discrimination of proteins as enzymes, and the recognition of DNA binding proteins. In both cases, our proposed approach achieves better prediction performance than existing methods. Second, we propose two novel semantic similarity measures for proteins based on the gene ontology. The first measure directly works on the gene ontology by combining the pairwise semantic similarity scores between sets of annotating terms for a pair of input proteins. The second measure estimates protein semantic similarity using a shortest path graph kernel to take advantage of the rich semantic knowledge contained within ontologies. Our comparison with other methods shows that our proposed semantic similarity measures are highly competitive and the latter one outperforms state-of-the-art methods. Furthermore, our two methods are intrinsic to the gene ontology, in the sense that they do not rely on external sources to calculate similarities.

A wide range of statistical and machine learning problems involve learning one or multiple latent functions, or properties thereof, from datasets. Examples include regression, classification, principal component analysis, optimisation, learning intensity functions of point processes and reinforcement learning to name but a few. For all these problems, positive semi-definite kernels (or simply kernels) provide a powerful tool for postulating flexible nonparametric hypothesis spaces over functions. Despite recent work on such kernel methods, parametric alternatives, such as deep neural networks, have been at the core of most artificial intelligence breakthroughs in recent years. In this thesis, both theoretical and methodological foundations are presented for constructing fully automated, scalable, and general-purpose kernel machines that perform very well over a wide range of input dimensions and sample sizes. This thesis aims to contribute towards bridging the gap between kernel methods and deep learning and to propose methods that have the advantage over deep learning in performing well on both small and large scale problems. In Part I we provide a gentle introduction to kernel methods, review recent work, identify remaining gaps and outline our contributions. In Part II we develop flexible and scalable Bayesian kernel methods in order to address gaps in methods capable of dealing with the special case of datasets exhibiting locally homogeneous patterns. We begin with two motivating applications. First we consider inferring the intensity function of an inhomogeneous point process in Chapter 2. This application is used to illustrate that often, by carefully adding some mild asymmetry in the dependency structure in Bayesian kernel methods, one may considerably scale-up inference while improving flexibility and accuracy. In Chapter 3 we propose a scalable scheme for online forecasting of time series and fully-online learning of related model parameters, under a kernel-based generative model that is provably sufficiently flexible. This application illustrates that, for one-dimensional input spaces, restricting the degree of differentiability of the latent function of interest may considerably speed-up inference without resorting to approximations and without any adverse effect on flexibility or accuracy. Chapter 4 generalizes these approaches and proposes a novel class of stochastic processes we refer to as string Gaussian processes (string GPs) that, when used as functional prior in a Bayesian nonparametric framework, allow for inference in linear time complexity and linear memory requirement, without resorting to approximations. More importantly, the corresponding inference scheme, which we derive in Chapter 5, also allows flexible learning of locally homogeneous patterns and automated learning of model complexity - that is automated learning of whether there are local patterns in the data in the first place, how much local patterns are present, and where they are located. In Part III we provide a broader discussion covering all types of patterns (homogeneous, locally homogeneous or heterogeneous patterns) and both Bayesian or frequentist kernel methods. In Chapter 6 we begin by discussing what properties a family of kernels should possess to enable fully automated kernel methods that are applicable to any type of datasets. In this chapter, we discuss a novel mathematical formalism for the notion of ‘general-purpose' families of kernels, and we argue that existing families of kernels are not general-purpose. In Chapter 7 we derive weak sufficient conditions for families of kernels to be general-purpose, and we exhibit tractable such families that enjoy a suitable parametrisation, that we refer to as generalized spectral kernels (GSKs). In Chapter 8 we provide a scalable inference scheme for automated kernel learning using general-purpose families of kernels. The proposed inference scheme scales linearly with the sample size and enables automated learning of nonstationarity and model complexity from the data, in virtually any kernel method. Finally, we conclude with a discussion in Chapter 9 where we show that deep learning can be regarded as a particular type of kernel learning method, and we discuss possible extensions in Chapter 10.

No description available.

The main objective of this study was to quantify changes in physical features of western Canadian wheat kernels caused by moisture increase using a machine vision system. Single wheat kernels of eight western Canadian wheat classes were conditioned to 12, 14, 16, 18, and 20% (wet basis) moisture content, one after another, using headspaces above various concentrations of potassium hydroxide (KOH) solutions which regulated relative humidity. A digital camera of 7.4 x 7.4 μm pixel resolution with an inter-line transfer charge-coupled device (CCD) image sensor was used to acquire images of individual kernels of all samples. A machine vision algorithm developed at the Canadian Wheat Board Centre for Grain Storage Research, University of Manitoba, was implemented to extract 49 morphological features from the wheat kernel images. Of the 49 morphological features, 24, 11, 7, 21, 26, 11, 17, and 9 features of Canada Western Red Spring, Canada Western Amber Durum, Canada Prairie Spring White, Canada Prairie Spring Red, Canada Western Extra Strong, Canada Western Red Winter, Canada Western Hard White Spring, and Canada Western Soft White Spring wheat kernels, respectively, were significantly (α=0.05) different as the moisture content increased from 12 to 20%. Generally the basic morphological features such as area, perimeter, major axis length, minor axis length, maximum radius, minimum radius, and mean radius were linearly increased with increase in moisture content. In all cases the moment and Fourier descriptor features decreased as moisture content increased from 12 to 20%.

The main objective of this study was to quantify changes in physical features of western Canadian wheat kernels caused by moisture increase using a machine vision system. Single wheat kernels of eight western Canadian wheat classes were conditioned to 12, 14, 16, 18, and 20% (wet basis) moisture content, one after another, using headspaces above various concentrations of potassium hydroxide (KOH) solutions which regulated relative humidity. A digital camera of 7.4 x 7.4 μm pixel resolution with an inter-line transfer charge-coupled device (CCD) image sensor was used to acquire images of individual kernels of all samples. A machine vision algorithm developed at the Canadian Wheat Board Centre for Grain Storage Research, University of Manitoba, was implemented to extract 49 morphological features from the wheat kernel images. Of the 49 morphological features, 24, 11, 7, 21, 26, 11, 17, and 9 features of Canada Western Red Spring, Canada Western Amber Durum, Canada Prairie Spring White, Canada Prairie Spring Red, Canada Western Extra Strong, Canada Western Red Winter, Canada Western Hard White Spring, and Canada Western Soft White Spring wheat kernels, respectively, were significantly (α=0.05) different as the moisture content increased from 12 to 20%. Generally the basic morphological features such as area, perimeter, major axis length, minor axis length, maximum radius, minimum radius, and mean radius were linearly increased with increase in moisture content. In all cases the moment and Fourier descriptor features decreased as moisture content increased from 12 to 20%.

Thesis (Master of Agriculture)--Cape Peninsula University of Technology, 2020 / Grain yield and its components are very important and complicated in barley (Hordeum vulgare L.) and are highly influenced by environmental factors and agronomic management practices. For 2018 growing season, a study was designed under rainfed conditions to evaluate the effects of nitrogen (N) fertilizer rate (0, 10, 20, 30 and 40 kg ha-1 of N) and planting density (120, 140, 160, and 180 to 200 seeds m-2) on the agronomic performance of three barley cultivars (Elim, Hessekwa and S16). A randomized complete block design with 3 replications was used. Combined analysis of variance showed significant (p<0.1) differences among cultivars, N rates and planting densities. The main objective of this study was to determine the effects of planting density and different fertilizer application strategies on barley grain yield and quality. The results showed that biggest increases on yield and yield components were observed at 180 seeds m-2 and 80kg ha-1 N rate. Higher N rates generally reduced kernel size. Kernel size was both increased and decreased by increasing planting density as well as N rate. Increasing planting density from 180 to 200 seeds m–2 generally provided slight reductions in grain N concentration and reduced kernel size. The three cultivars expressed a significant effect on kernel plumpness and N content of grain. The most beneficial agronomic practices for malting barley production in Western Cape were application of N fertilizer at optimum rate depending on cultivar, locality and rainfall and planting seeds at a rate of 160-180 seeds m-2 depending on cultivar. A planting density of 160-180 seeds m-2 at a rate of 80 kg N ha-1 is recommended for planting barley under dry land in the Western Cape.

Hidden undesired functionality is an unavoidable reality in any complex hardware or software component. Undesired functionality — deliberately introduced Trojan horses or accidentally introduced bugs — in any component of a system can be exploited by attackers to exert control over the system. This poses a serious security risk to systems — especially in the ever growing number of systems based on networks of computers. The approach adopted in this dissertation to secure systems seeks immunity from hidden functionality. Specifcally, if a minimal trusted computing base (TCB) for any system can be identifed, and if we can eliminate hidden functionality in the TCB, all desired assurances regarding the operation of the system can be guaranteed. More specifcally, the desired assurances are guaranteed even if undesired functionality may exist in every component of the system outside the TCB. A broad goal of this dissertation is to characterize the TCB for various systems as a set of functions executed by a trusted security kernel. Some constraints are deliberately imposed on the security kernel functionality to reduce the risk of hidden functionality inside the security kernel. In the security model adopted in this dissertation, any system is seen as an interconnection of subsystems, where each subsystem is associated with a security kernel. The security kernel for a subsystem performs only the bare minimal tasks required to assure the integrity of the tasks performed by the subsystem. Even while the security kernel functionality may be different for each system/subsystem, it is essential to identify reusable components of the functionality that are suitable for a wide range of systems. The contribution of the research is a versatile data-structure — Ordered Merkle Tree (OMT), which can act as the reusable component of various security kernels. The utility of OMT is illustrated by designing security kernels for subsystems participating in, 1) a remote fle storage system, 2) a generic content distribution system, 3) generic look-up servers, 4) mobile ad-hoc networks and 5) the Internet’s routing infrastructure based on the border gateway protocol (BGP).