The protein fold recognition problem is one of the important topics in biophysics. It is believed that the primary structure of a protein is helpful to drawing its three-dimensional (3D) structure. Given a target protein (a sequence of amino acids), the protein fold recognition problem is to decide which fold group of some protein structure database the target protein belongs to. Since more than two fold groups are to be located in this problem, it is a multi-class classification problem. Recently, many researchers have solved this problem by using the popular machine learning tools, such as neural networks (NN) and support vector machines (SVM). In this thesis, we use the SVM tool to solve this problem. Our strategy is to find out the effective features which can be used as an efficient guide to the classification problem. We build the feature preference table to help us to find out effective feature combinations quickly. We take 27 well-known fold groups in SCOP (Structural Classification of Proteins) as our data set. Our experimental results show that our method achieves the overall prediction accuracy of 61.4%, which is better than the previous method (56.5%). With the same feature combinations, our prediction accuracy is also higher than the previous results. These results show that our method is indeed effective for the fold recognition problem.

This thesis proposes an approach to segmenting and identifying mixed-language speech. Automatic LID can be divided into four steps, feature extraction, segmentation, segment clustering, and re-labeling. In feature extraction, we compare the group delay feature (GDF) with MFCC feature. Unlike the traditional feature from Fourier trans-form magnitude, GDF uses the phase spectrum. In segmentation, we compare delta Bayesian information criterion (delta-BIC) with support vector machines (SVMs). A delta-BIC is applied to segment the input speech utterance into a sequence of lan-guage-dependent segments using acoustic features. The segments are clustered using the K-means algorithm. Finally, re-labeling is used to determine the language of the clusters. SVMs proceed to segment and identify automatically after model training. Considering the effect of the accent issue, we use the corpus English Across Taiwan (EAT) to perform our system. The experimental results show that the system can reach 78.13% in the frame hit rate under the baseline 57.77%.

Wrinkling caused in wearing and laundry procedures is one of the most important performance properties of a fabric. Visual examination performed by trained experts is a routine wrinkle evaluation method in textile industry, however, this subjective evaluation is time-consuming. The need for objective, automatic and efficient methods of wrinkle evaluation has been increasing remarkably in recent years. In the present thesis, a wavelet transform based imaging analysis method was developed to measure the 2D fabric surface data captured by an infrared imaging system. After decomposing the fabric image by the Haar wavelet transform algorithm, five parameters were defined based on modified wavelet coefficients to describe wrinkling features, such as orientation, hardness, density and contrast. The wrinkle parameters provide useful information for textile, appliance, and detergent manufactures who study wrinkling behaviors of fabrics. A Support-Vector-Machine based classification scheme was developed for automatic wrinkle rating. Both linear kernel and radial-basis-function (RBF) kernel functions were used to achieve a higher rating accuracy. The effectiveness of this evaluation method was tested by 300 images of five selected fabric types with different fiber contents, weave structures, colors and laundering cycles. The results show agreement between the proposed wavelet-based automatic assessment and experts’ visual ratings. / text

No description available.

Classifying pipe cracks by size from their pulse-echo ultrasonic signal is difficult but highly significant for the defect evaluation required in pipe testing and maintenance decision making. For this thesis, a binary Support Vector Machine (SVM) classifier, which divides pipe cracks into two categories: large and small, was developed using collected ultrasonic signals. To improve the performance of this SVM classifier in terms of reducing test errors, we first combined the Sequential Backward Selection and Sequential Forward Selection schemes for input feature reduction. Secondly, we used the data dependent kernel instead of the Gaussian kernel as the kernel function in the SVM classifier. Thirdly, as it is time-consuming to use the classic grid-search method for parameter selection of SVM, this work proposes a Kernel Fisher Discriminant Ratio (KFD Ratio) which makes it possible to more quickly select parameters for the SVM classifier.

Made available in DSpace on 2016-08-29T15:33:14Z (GMT). No. of bitstreams: 1 tese_4163_.pdf: 479699 bytes, checksum: 04f01a137084c0859b4494de6db8b3ac (MD5) Previous issue date: 2011-02-23 / Classificadores do tipo máquina de vetores de suporte (SVM) são atualmente considerados uma das técnicas mais poderosas para se resolver problemas de classificação com duas classes. Para aumentar o desempenho alcançado por classificadores SVM individuais, uma abordagem bem estabelecida é usar uma combinação de SVMs, a qual corresponde a um conjunto de classificadores SVMs que são, simultaneamente, individualmente precisos e coletivamente divergentes em suas decisões. Este trabalho propõe uma abordagem para se criar combinações de SVMs, baseada em um processo de três estágios. Inicialmente, são usadas execuções complementares de uma busca baseada em algoritmos genéticos (GEFS), com o objetivo de investigar globalmente o espaço de características para definir um conjunto de subconjuntos de características. Em seguida, para cada um desses subconjuntos de características definidos, uma SVM que usa parâmetros otimizados é construída. Por fim, é empregada uma busca local com o objetivo de selecionar um subconjunto otimizado dessas SVMs, e assim formar a combinação de SVMs que é finalmente produzida. Os experimentos foram realizados num contexto de detecção de defeitos em máquinas industriais. Foram usados 2000 exemplos de sinais de vibração de moto bombas instaladas em plataformas de petróleo. Os experimentos realizados mostram que o método proposto para se criar combinação de SVMs apresentou um desempenho superior em comparação a outras abordagens de classificação bem estabelecidas.

A Thesis submitted to the Faculty of Science, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy, 2017 / The utility of lectures is directly related to the engagement of the students therein. To ensure the value of lectures, one needs to be certain that they are engaging to students. In small classes experienced lecturers develop an intuition of how engaged the class is as a whole and can then react appropriately to remedy the situation through various strategies such as breaks or changes in style, pace and content. As both the number of students and size of the venue grow, this type of contingent teaching becomes increasingly difficult and less precise. Furthermore, relying on intuition alone gives no way to recall and analyse previous classes or to objectively investigate trends over time. To address these problems this thesis presents the WITS INTELLIGENT TEACHING SYSTEM (WITS) to highlight disengaged students during class. A web-based, mobile application called Engage was developed to try elicit anonymous engagement information directly from students. The majority of students were unwilling or unable to self-report their engagement levels during class. This stems from a number of cultural and practical issues related to social display rules, unreliable internet connections, data costs, and distractions. This result highlights the need for a non-intrusive system that does not require the active participation of students. A nonintrusive, computer vision and machine learning based approach is therefore proposed. To support the development thereof, a labelled video dataset of students was built by recording a number of first year lectures. Students were labelled across a number of affects – including boredom, frustration, confusion, and fatigue – but poor inter-rater reliability meant that these labels could not be used as ground truth. Based on manual coding methods identified in the literature, a number of actions, gestures, and postures were identified as proxies of behavioural engagement. These proxies are then used in an observational checklist to mark students as engaged or not. A Support Vector Machine (SVM) was trained on Histograms of Oriented Gradients (HOG) to classify the students based on the identified behaviours. The results suggest a high temporal correlation of a single subject’s video frames. This leads to extremely high accuracies on seen subjects. However, this approach generalised poorly to unseen subjects and more careful feature engineering is required. The use of Convolutional Neural Networks (CNNs) improved the classification accuracy substantially, both over a single subject and when generalising to unseen subjects. While more computationally expensive than the SVM, the CNN approach lends itself to parallelism using Graphics Processing Units (GPUs). With GPU hardware acceleration, the system is able to run in near real-time and with further optimisations a real-time classifier is feasible. The classifier provides engagement values, which can be displayed to the lecturer live during class. This information is displayed as an Interest Map which highlights spatial areas of disengagement. The lecturer can then make informed decisions about how to progress with the class, what teaching styles to employ, and on which students to focus. An Interest Map was presented to lecturers and professors at the University of the Witwatersrand yielding 131 responses. The vast majority of respondents indicated that they would like to receive live engagement feedback during class, that they found the Interest Map an intuitive visualisation tool, and that they would be interested in using such technology. Contributions of this thesis include the development of a labelled video dataset; the development of a web based system to allow students to self-report engagement; the development of cross-platform, open-source software for spatial, action and affect labelling; the application of Histogram of Oriented Gradient based Support Vector Machines, and Deep Convolutional Neural Networks to classify this data; the development of an Interest Map to intuitively display engagement information to presenters; and finally an analysis of acceptance of such a system by educators. / XL2017

Data generated by complex, computational models can provide highly accurate predictions of hydrological and hydrodynamic data in multiple dimensions. Unfortunately, however, for large data sets, running these models is often timeconsuming and computationally expensive. Thus, finding a way to reduce the running time of these models, while still producing comparable results, is of notable interest. The Inference Engine is a proposed system for doing just this. It takes previously generated model data and uses them to predict additional data. Its performance, both accuracy and running time, has been compared to the performance of the actual models, in increasingly difficult data prediction tasks, and it is able, with sufficient accuracy, to quickly predict unknown model data.

In this thesis, a support vector machine (SVM) is used to develop a model to predict Arias Intensity. Arias Intensity is a measure of the strength of ground motions that considers both the amplitude and the duration of ground motions. In this research, a subset of the database from the “Next Generation and the duration of Ground-Motion Attenuation Models” project was used as the training data. The data includes 3525 ground motion records from 175 earthquakes. This research provides the assessment of historical earthquakes using arias intensity data. Support vector machine uses a Kernel function to transform the data into a high dimensional space where relationships between the variables can be efficiently described using simpler models. In this research, after testing several kernel functions, a Gaussian Kernel was selected for the predictive model. The resulting model uses magnitude, epicentral distance, and the shear wave velocity as the predictor of Arias Intensity.

This thesis studies the problem of frequentist model averaging over a set of multiple $\epsilon$-support vector regression (SVR) models, where the support vector machine (SVM) algorithm was extended to function estimation involving continuous targets, instead of categorical ones. By assigning weights to a set of candidate models instead of selecting the least misspecified one, model averaging presents a strong alternative to model selection for tackling model uncertainty. Not only do we describe the construction of smoothed BIC/AIC model averaging weights, but we also propose a Mallows model averaging procedure which selects model weights by minimizing Mallows' criterion. We conduct two studies where the set of candidate models can either include or not include the true model by making use of simulated random samples obtained from different data-generating processes of analytic form. In terms of mean squared error, we demonstrate that our proposed method outperforms other model averaging and model selection methods that were tested, and the gain is more substantial for smaller sample sizes with larger signal-to-noise ratios. / Thesis / Master of Science (MSc)