Research output can be seen as the result of research and study which is done by an institutions' research community - usually in publicised form such as an article or book. The NWU-IR drives to raise the visibility of the outputs both locally and internationally to create further awareness of what is being done by the NWU.

Collections in this community

Recent Submissions

  • Tracking translation invariance in CNNs 

    Myburgh, Johannes C.; Mouton, Coenraad; Davel, Marelie H. (Southern African Conference for Artificial Intelligence Research, 2020)
    Although Convolutional Neural Networks (CNNs) are widely used, their translation invariance (ability to deal with translated inputs) is still subject to some controversy. We explore this question using translation-sensitivity ...
  • Stride and translation invariance in CNNs 

    Mouton, Coenraad; Myburgh, Johannes C.; Davel, Marelie H. (Southern African Conference for Artificial Intelligence Research, 2020)
    Convolutional Neural Networks have become the standard for image classification tasks, however, these architectures are not invariant to translations of the input image. This lack of invariance is attributed to the use of ...
  • Pairwise networks for feature ranking of a geomagnetic storm model 

    Beukes, Jacques Pieter; Davel, Marelie Hattingh; Lotz, Stefan (South African Institute of Computer Scientists and Information Technologists, 2020)
    Feedforward neural networks provide the basis for complex regression models that produce accurate predictions in a variety of applications. However, they generally do not explicitly provide any information about the utility ...
  • Using summary layers to probe neural network behaviour 

    Davel, Marelie Hattingh (South African Institute of Computer Scientists and Information Technologists, 2020)
    No framework exists that can explain and predict the generalisation ability of deep neural networks in general circumstances. In fact, this question has not been answered for some of the least complicated of neural network ...
  • Benign interpolation of noise in deep learning 

    Davel, Marelie Hattingh; Barnard, Etienne; Theunissen, Marthinus Wilhelmus (South African Institute of Computer Scientists and Information Technologists, 2020)
    The understanding of generalisation in machine learning is in a state of flux, in part due to the ability of deep learning models to interpolate noisy training data and still perform appropriately on out-of-sample data, ...
  • Pre-interpolation loss behavior in neural networks 

    Venter, Arthur Edgar William; Theunissen, Marthinus Wilhelm; Davel, Marelie Hattingh (Springer, 2020)
    When training neural networks as classifiers, it is common to observe an increase in average test loss while still maintaining or improving the overall classification accuracy on the same dataset. In spite of the ubiquity ...
  • The South African directory enquiries (SADE) name corpus 

    Thirion, Jan Willem Frederick; Van Heerden, Charl Johannes; Giwa, Oluwapelumi; Davel, Marelie Hattingh (Springer, 2020)
    We present the design and development of a South African directory enquiries (DE) corpus. It contains audio and orthographic transcriptions of a wide range of South African names produced by first language speakers of four ...
  • Optimising word embeddings for recognised multilingual speech 

    Barnard, Etienne; Heyns, Nuette (Southern African Conference for Artificial Intelligence Research, 2020)
    Word embeddings are widely used in natural language processing (NLP) tasks. Most work on word embeddings focuses on monolingual languages with large available datasets. For embeddings to be useful in a multilingual ...
  • Classifying recognised speech with deep neural networks 

    Strydom, Rhyno A; Barnard, Etienne (Southern African Conference for Artificial Intelligence Research, 2020)
    We investigate whether word embeddings using deep neural networks can assist in the analysis of text produced by a speechrecognition system. In particular, we develop algorithms to identify which words are incorrectly ...
  • Exploring neural network training dynamics through binary node activations 

    Haasbroek, Daniël G.; Davel, Marelie H. (Southern African Conference for Artificial Intelligence Research, 2020)
    Each node in a neural network is trained to activate for a specific region in the input domain. Any training samples that fall within this domain are therefore implicitly clustered together. Recent work has highlighted ...
  • Using a meta-model to compensate for training-evaluation mismatches 

    Lamprecht, Dylan; Barnard, Etienne (Southern African Conference for Artificial Intelligence Research, 2020)
    One of the fundamental assumptions of machine learning is that learnt models are applied to data that is identically distributed to the training data. This assumption is often not realistic: for example, data collected ...
  • Convolution algebra for extended Feller convolution 

    Lee, Wha-Suck; Le Roux, Christiaan (Springer, 2020)
    We apply the recently introduced framework of admissible homomorphisms in the form of a convolution algebra of C2-valued admissible homomorphisms to handle two-dimensional uni-directional homogeneous stochastic kernels. ...
  • Comments on: Tests for multivariate normalit: a critical review with emphasis on weighted L2 -statistics 

    Meintanis, Simos G. (Springer, 2020)
    We discuss extension of the BHEP test to more general families of distributions
  • Comparison of body mass index and fat percentage criteria classification of 7-13 year-old rural boys in South Africa 

    Van Gent, Maya; Pienaar, Anita; Noorbhai, Habib (BMC, 2020)
    Abstract Background: The aim of this paper was to investigate whether BMI and fat percentage classification criteria, would classify a sample of 7–13 year old boys from a rural background in similar nutritional ...
  • Ten research questions to support South Africa’s Inland Fisheries Policy 

    Weyl, O.L.F.; Smit, N.J.; Wepener, V.; Barkhuizen, L.; Christison, K. (Taylor & Francis, 2020)
    South Africa is in the process of developing a National Freshwater (Inland) Wild Capture Fisheries Policy. A properly focused research strategy is essential to guide the policy development process, and thus a dedicated ...
  • Chitosan composite biomaterials for bone tissue engineering: a review 

    Fourie, Jaundrie; Du Preez, Louis; Taute, Francois; De Beer, Deon (Springer, 2020)
    The bone is a highly dynamic tissue with the remarkable ability to remodel and is in a continuous cycle of resorption and renewal as a result of internal mediators and external mechanical demands. Researchers have doubled ...
  • Genetic fuzzy rule extraction for optimised sizing and control of hybrid renewable energy hydrogen systems 

    Human, G.; Van Schoor, G.; Uren, K.R. (Elsevier, 2020)
    A major challenge related to the design of a hybrid renewable energy hydrogen system is which energy sources to include and at what capacity, for regionally different potentials of renewable energy and hydrogen demand. In ...
  • Prevalence of dyslipidaemia among type 2 diabetes mellitus patients in the Western Cape, South Africa 

    Omodanisi, Elizabeth I.; Ntwampe, Seteno K.O.; Tomose, Yibanathi; Okeleye, Benjamin I.; Aboua, Yapo G. (MDPI, 2020)
    Dyslipidaemia, an irregular aggregate of lipids in the blood is common in diabetes and cardiovascular disease sufferers. A cross-sectional study on the prevalence of dyslipidaemia was performed among type 2 diabetes mellitus ...
  • Methylene blue analogues: in vitro antimicrobial minimum inhibitory concentrations and in silico pharmacophore modelling 

    Thesnaar, Louis; Bezuidenhout, Jaco J.; Petzer, Anél; Petzer, Jacobus P.; Cloete, Theunis T. (Elsevier, 2020)
    It has been shown that methylene blue has antimicrobial properties although few studies have determined its minimum inhibitory concentration (MIC), which is the gold standard used to measure antimicrobial activity. The ...

View more