The use of quadratic discriminant analysis (QDA) or its regularized version (RQDA) for classification is often not recommended, due to its well-acknowledged high sensitivity to the estimation noise of the covariance matrix. This becomes all the more the case in unbalanced data settings for which it has been found that R-QDA becomes equivalent to the classifier that assigns all observations to the same class. In this paper, we propose an improved R-QDA that is based on the use of two regularization parameters and a modified bias, properly chosen to avoid inappropriate behaviors of R-QDA in unbalanced settings and to ensure the best possible classification performance. The design of the proposed classifier builds on a refined asymptotic analysis of its performance when the number of samples and that of features grow large simultaneously, which allows to cope efficiently with the high-dimensionality frequently met within the big data paradigm. The performance of the proposed classifier is assessed on both real and synthetic data sets and was shown to be much higher than what one would expect from a traditional R-QDA.

In today’s world, we are always trying to find new ways to advance. This era has given rise to a global, distributed workforce since technology has allowed people to access and communicate with individuals all over the world. With the rise of remote workers, the need for quality communication tools has risen significantly. These communication tools come in many forms, and web-conference apps are among the most prominent for the task. Developing a system to automatically summarize the web-conference will save companies time and money, leading to more efficient meetings. Current approaches to summarizing multi-speaker web-conferences tend to yield poor or incoherent results, since conversations do not flow in the same manner that monologues or well-structured articles do. This thesis proposes a system of flags used to extract information from sentences, where the flags are fed into Machine Learning models to determine the importance of the the sentence with which they are associated. The system of flags shows promise for multi-speaker conference summaries.

With the increase in scholarly articles published every day, the need for an automated systematic exploratory literature review tool is rising. With the advance in Text Mining and Machine Learning methods, such data exploratory tools are researched and developed in every scientific domain. This research aims at finding the best keyphrase extraction algorithm and topic modeling algorithm that is going to be the foundation and main component of a tool that will aid in Systematic Literature Review. Based on experimentation on a set of highly relevant scholarly articles published in the domain of food science, two graph-based keyphrase extraction algorithms, TopicalPageRank and PositionRank were picked as the best two algorithms among 9 keyphrase extraction algorithms for picking domain-specific keywords. Among the two topic modeling algorithms, Latent Dirichlet Assignment (LDA) and Non-zero Matrix Factorization (NMF), documents chosen in this research were best classified into suitable topics by the NMF method validated by a domain expert. This research lays the framework for a faster tool development for Systematic Literature Review.

Malaria is a terrible disease caused by a protozoan parasite within the Plasmodium genus, claiming the lives of hundreds of thousands of people yearly, the majority of whom are children under the age of five. Of the five species of Plasmodium causing malaria in humans, P. falciparum is responsible for most of the death toll. An increase in malaria cases was detected between the years 2016 to 2017 according to the World Malaria Report of 2017, despite control efforts. The rapid development of resistance within P. falciparum against antimalarials has led to the use of artemisinin combinational therapy as the current gold standard for malaria treatment. Yet decreased parasite clearance demonstrates that using combination therapy is insufficient in maintaining current antimalarials’ effectiveness against these resistant parasites. Hence, novel compounds with a mode of action (MoA) different than current antimalarials are required. Though phenotypic screening has delivered thousands of promising hit compounds, hit-to-lead optimisation is still one of the rate-limiting steps in pre-clinical antimalarial drug development. While knowing the exact target or MoA is not required to progress a compound in a medicinal chemistry program, identifying the MoA early can accelerate hit prioritization, hit-to-lead optimisation and preclinical combination studies in malaria research. In this study, we assessed machine learning (ML) approaches for their ability to stratify antimalarials based on transcriptional responses associated with the treatments. From our results, we conclude that it is possible to identify biomarkers from the transcriptional responses that define the MoA of compounds. Moreover, only a limited set of 50 genes was required to build a ML model that can stratify compounds with similar MoA with a classification accuracy of 76.6 ± 6.4%. These biomarkers will help stratify new compounds with similar MoA to those already defined with our strategy. Additionally, the biomarkers can also be used to monitor if the MoA of a compound has changed during hit-to-lead optimisation. This work will contribute to accelerating antimalarial drug discovery during the hit-to-lead optimisation phase and help the identification of compounds with novel MoA. / Dissertation (MSc)--University of Pretoria, 2019. / Biochemistry / MSc / Unrestricted

The next-generation (6G) networks promise to provide extended 5G capabilities with enhanced performance at high data rates, low latency, low energy consumption, and rapid adaptation. 6G networks are also expected to support the unprecedented Internet of Everything (IoE) scenarios with highly diverse requirements. With the emerging applications of autonomous driving, virtual reality, and mobile computing, achieving better performance and fulfilling the diverse requirements of 6G networks are becoming increasingly difficult due to the rapid proliferation of wireless data and heterogeneous network structures. In this regard, learning-based algorithms are naturally powerful tools to deal with the numerous data and are expected to impact the evolution of communication networks. This thesis employed learning-based approaches to enhance the performance and fulfill the diverse requirements of the next-generation intelligent networks under various network structures. Specifically, we design the trajectory of the unmanned aerial vehicle (UAV) to provide energy-efficient, high data rate, and fair service for the Internet of things (IoT) networks by employing on/off-policy reinforcement learning (RL). Thereafter, we applied a deep RL-based approach for heterogeneous traffic offloading in the space-air-ground integrated network (SAGIN) to cover the co-existing requirements of ultra-reliable low-latency communication (URLLC) traffic and enhanced mobile broadband (eMBB) traffic. Precise traffic prediction can significantly improve the performance of 6G networks in terms of intelligent network operations, such as predictive network configuration control, traffic offloading, and communication resource allocation. Therefore, we investigate the wireless traffic prediction problem in edge networks by applying a federated meta-learning approach. Lastly, we design an importance-oriented clustering-based high quality of service (QoS) system with software-defined networking (SDN) by adopting unsupervised learning.

This work presents an enhancement to the classification tree algorithm which forms the basis for Random Forests. Differently from the classical tree-based methods that focus on one variable at a time to separate the observations, the new algorithm performs the search for the best split in two-dimensional space using a linear combination of variables. Besides the classification, the method can be used to determine variables interaction and perform feature extraction. Theoretical investigations and numerical simulations were used to analyze the properties and performance of the new approach. Comparison with other popular classification methods was performed using simulated and real data examples. The algorithm was implemented as an extension package for the statistical computing environment R and is available for free download under the GNU General Public License.

Diese Arbeit beschäftigt sich mit einem realen Fall der Chargenproduktion aus der pharmazeutischen Industrie. Das in der Untersuchung betrachtete Problem liegt im Bereich der Optimierung der Chargenqualität und der Minimierung des Ausschusses unter der Gegebenheit, dass die entsprechenden Qualitätsparameter im Unternehmenssteuerungssystem nicht gemessen werden. Die in dieser Arbeit vorgeschlagene Technik führt ein virtuelles Qualitätskriterium ein, das für jede der Chargen angewendet wird, basierend auf dem beschränkten Wissen der Anwender, welche Charge als optimale Charge (auch Golden Batch bezeichnet) betrachtet werden kann und somit als Referenz für die aktuell in Produktion befindliche Charge verwendet werden kann. Zu diesem Zweck verwenden wir das klassische integrale Leistungskriterium, das in der Theorie der optimalen Steuerung dynamischer Systeme weit verbreitet ist, um zu messen, wie weit der aktuelle Zustand des Systems vom 'optimalen' Punkt entfernt ist. Mit Hilfe der beschriebenen Technologie, die aus der genannten Nachbardisziplin stammt, waren wir in der Lage, die Qualität jeder Charge als ein kontinuierliches Messverhältnis zu quantifizieren, was uns erlaubte, mehrere effiziente kontinuierliche Analysetechniken für diesen anfänglichen Chargenproduktionsfall zu verwenden.

Indiana University-Purdue University Indianapolis (IUPUI) / Recently witnessed a great popularity of unobtrusive Visible Light Communication (VLC) using screen-camera channels. They overcomes the inherent drawbacks of traditional approaches based on coded images like bar codes. One popular unobtrusive method is the utilizing of alpha channel or color channels to encode bits into the pixel translucency or color intensity changes with over-the-shelf smart devices. Specifically, Uber-in-light proves to be an successful model encoding data into the color intensity changes that only requires over-the-shelf devices. However, Uber-in-light only exploit Multi Frequency Shift Keying (MFSK), which limits the overall throughput of the system since each data segment is only 3-digit long. Motivated by some previous works like Inframe++ or Uber-in-light, in this thesis, we proposes a new VLC model encoding data into color intensity changes on red and blue channels of video frames. Multi-Phase-Shift-Keying (MPSK) along with MFSK are used to match 4-digit and 5-digit long data segments to specific transmission frequencies and phases. To ensure the transmission accuracy, a modified correlation-based demodulation method and two learning-based methods using SVM and Random Forest are also developed.

No description available.

Data assimilation of satellite-based observations of hydrological variables with full numerical physics models can be used to downscale these observations from coarse to high resolution to improve microwave sensor-based soil moisture observations. Moreover, assimilation can also be used to predict related hydrological variables, e.g., precipitation products can be assimilated in a land information system to estimate soil moisture. High quality spatio-temporal observations of these processes are vital for a successful assimilation which in turn needs a detailed analysis and improvement. In this research, pattern recognition and adaptive signal processing methods are developed for the spatio-temporal analysis and enhancement of soil moisture and precipitation datasets. These methods are applied to accomplish the following tasks: (i) a consistency analysis of level-3 soil moisture data from the Advanced Microwave Scanning Radiometer – EOS (AMSR-E) against in-situ soil moisture measurements from the USDA Soil Climate Analysis Network (SCAN). This method performs a consistency assessment of the entire time series in relation to others and provides a spatial distribution of consistency levels. The methodology is based on a combination of wavelet-based feature extraction and oneclass support vector machines (SVM) classifier. Spatial distribution of consistency levels are presented as consistency maps for a region, including the states of Mississippi, Arkansas, and Louisiana. These results are well correlated with the spatial distributions of average soil moisture, and the cumulative counts of dense vegetation; (ii) a modified singular spectral analysis based interpolation scheme is developed and validated on a few geophysical data products including GODAE’s high resolution sea surface temperature (GHRSST). This method is later employed to fill the systematic gaps in level-3 AMSR-E soil moisture dataset; (iii) a combination of artificial neural networks and vector space transformation function is used to fuse several high resolution precipitation products (HRPP). The final merged product is statistically superior to any of the individual datasets over a seasonal period. The results have been tested against ground based measurements of rainfall over our study area and average accuracies obtained are 85% in the summer and 55% in the winter 2007.