In this thesis, many machine learning algorithms were applied to electrocardiogram (ECG), spectral analysis, and Field Programmable Gate Arrays (FPGAs). In ECG, QRS complexes are useful for measuring the heart rate and for the segmentation of ECG signals. QRS complexes were detected using WaveletCNN Autoencoder filters and ConvLSTM detectors. The WaveletCNN Autoencoders filters the ECG signals using the wavelet filters, while the ConvLSTM detects the spatial temporal patterns of the QRS complexes. For the spectral analysis topic, the detection of chemical compounds using spectral analysis is useful for identifying unknown substances. However, spectral analysis algorithms require vast amounts of data. To solve this problem, B-spline neural networks were developed for the generation of infrared and ultraviolet/visible spectras. This allowed for the generation of large training datasets from a few experimental measurements. Graphical Processing Units (GPUs) are good for training and testing neural networks. However, using multiple GPUs together is hard because PCIe bus is not suited for scattering operations and reduce operations. FPGAs are more flexible as they can be arranged in a mesh or toroid or hypercube configuration on the PCB. These configurations provide higher data throughput and results in faster computations. A general neural network framework was written in VHDL for Xilinx FPGAs. It allows for any neural network to be trained or tested on FPGAs. / Graduate

The demand for accurate localization under indoor environments has increased dramatically in recent years. To be cost-effective, most of the localization solutions are based on the WiFi signals, utilizing the pervasive deployment of WiFi infrastructure and availability of the WiFi enabled mobile devices. In this thesis, we develop completed indoor localization solutions based on WiFi fingerprinting and machine learning approaches with two types of WiFi fingerprints including received signal strength indicator (RSSI) and channel state information (CSI). Starting from the low complexity algorithm, we propose a soft range limited K nearest neighbours (SRL-KNN) to address spatial ambiguity and the fluctuation of WiFi signals. SRL-KNN exploits RSSI and scales the fingerprint distance by a range factor related to the physical distance between the user’s previous position and the reference location in the database. Although utilizing the prior locations, SRL-KNN does not require knowledge of the exact moving speed and direction of the user. Besides, the idea of the soft range limiting factor can be applied to all of the existed probabilistic methods, i.e., parametric and nonparametric methods, to improve their performances. A semi-sequential short term memory step is proposed to add to the existed probabilistic methods to reduce their spatial ambiguity of fingerprints and boost significantly their localization accuracy. In the following research phase, instead of locating user's position one at a time as in the cases of conventional algorithms, our recurrent neuron networks (RNNs) solution aims at trajectory positioning and takes into account of the relation among RSSI measurements in a trajectory. The results using different types of RNN including vanilla RNN, long short-term memory (LSTM), gated recurrent unit (GRU) and bidirectional LSTM (BiLSTM) are presented. Next, the problem of localization using only one single router is analysed. CSI information will be adopted along with RSSI to enhance the localization accuracy. Each of the reference point (RP) is presented by a group of CSI measurements from several WiFi subcarriers which we call CSI images. The combination of convolutional neural network (CNN) and LSTM model is proposed. CNN extracts the useful information from several CSI values (CSI images), and then LSTM will exploit this information in sequential timesteps to determine the user's location. Finally, a fully practical passive indoor localization is proposed. Most of the conventional methods rely on the collected WiFi signal on the mobile devices (active information), which requires a dedicated software to be installed. Different from them, we leverage the received data of the routers (passive information) to locate the position of the user. The localization accuracy is investigated through experiments with several phones, e.g., Nexus 5, Samsung, Iphone and HTC, in hundreds of testing locations. The experimental results demonstrate that our proposed localization scheme achieves an average localization error of around 1.5 m when the phone is in idle mode, and approximately 1 m when it actively transmits data. / Graduate

No description available.

Recommendation systems have significantly influenced user experiences across various industries, yet their application in the insurance sector remains relatively unexplored. This thesis focuses on developing a car insurance recommendation system that implements a `consumers like you' feature. The study initially employs a clustering-based recommendation system due to missing labels in an offline environment. However, challenges emerge, such as determining the optimal number of clusters and managing complex data. Additionally, the inability to effectively update based on feedback and lower predictive performance compared to supervised methods necessitated exploring supervised alternatives. In response, this thesis proposes a methodology where the unsupervised approach simulates consumer behavior in an offline environment. Supervised alternatives are pre-trained on the clustering-based system to replicate it and come with the ability to be fine-tuned based on live traffic. Three supervised alternatives — KNN, XGBoost, and a neural network — are developed and compared. Given the supervised recommendation system adaptability based on feedback, supervised methods can provide more accurate, personalized recommendations in the insurance domain. The XGBoost and neural network-based recommendation systems were able to replicate the unsupervised approach, and their expressive power makes them valid candidate models to further evaluate on live traffic. The thesis concludes with the potential to both improve and adapt these recommendation systems to other insurance types, marking a significant step toward more personalized, user-friendly insurance services.

No description available.

The k-nearest neighbor (k-NN) pattern classifier is a simple yet effective learner. However, it has a few drawbacks, one of which is the large model size. There are a number of algorithms that are able to condense the model size of the k-NN classifier at the expense of accuracy. Boosting is therefore desirable for increasing the accuracy of these condensed models. Unfortunately, there does not exist a boosting algorithm that works well with k-NN directly. We present a direct boosting algorithm for the k-NN classifier that creates an ensemble of models with locally modified distance weighting. An empirical study conducted on 10 standard databases from the UCI repository shows that this new Boosted k-NN algorithm has increased generalization accuracy in the majority of the datasets and never performs worse than standard k-NN.

No description available.

No description available.

Competency-based medical education (CBME) is a paradigm of assessing resident performance through well-defined tasks, objectives and milestones. A large number of data points are generated during a five-year period as a resident accomplishes the assigned tasks. However, no tool support exists to process this data for early identification of a resident-at-risk failing to achieve future milestones. In this thesis, the implementation of CBME at McMaster's Royal College Emergency Medicine residency program was studied and the development of a machine learning algorithm (MLA) to identify patterns in resident performance was reported. The adaptivity of multiple MLAs to build a tool support for monitoring residents' progress and flagging those who are in most need of assistance in the context of emergency medicine education was evaluated. / Thesis / Master of Science (MSc)

This study evaluated the ability of segment-based classification paired with non-parametric methods (CART and kNN) to classify a chronosequence of Landsat TM/ETM+ imagery spanning from 1992 to 2002 within the state of Rondônia, Brazil. Pixel-based classification was also implemented for comparison. Interannual multitemporal composites were used in each classification in an attempt to increase the separation of primary forest, cleared, and re-vegetated classes within a given year. The kNN and CART classification methods, with the integration of multitemporal data, performed equally well with overall accuracies ranging from 77% to 91%. Pixel-based CART classification, although not different in terms of mean or median overall accuracy, did have significantly lower variability than all other techniques (3.2% vs. an average of 13.2%), and thus provided more consistent results. Segmentation did not improve classification success over pixel-based methods and was therefore an unnecessary processing step with the used dataset. Through the appropriate band selection methods of the respective non-parametric classifiers, multitemporal bands were chosen in 38 of the 44 total classifications, strongly suggesting the utility of interannual multitemporal data for the separation of cleared, re-vegetated, and primary forest classes. The separation of the primary forest class from the cleared and re-vegetated classes was particularly successful and may be a possible result of the incorporation of multitemporal data. The land cover maps from this study allow for an accurate annualized analysis of land cover and can be coupled with household data to gain a better understanding of landscape change in the region. / Master of Science