Students
Home » Engineering & Built Environment » Centers | Units | Institutes | Stations » Institute for Intelligent Systems »Peter O. Olukanmi
Research Title: Improved algorithms for k-means clustering
Qualification Enrolled for: PhD
Brief of Research Work: My current work fits into a bigger vision of meeting demands of this ‘data age’ for robust data methods that can handle ‘dirty’ and/or ‘large’ datasets. It focuses on a widely studied computational problem: clustering. k-means, arguably the most popular clustering method, is plagued with drawbacks such as poor scalability; sensitivity to initialization and outliers; assumed knowledge of cluster count; and production of local rather than global optimum. My work seeks to advance the state of the art in addressing these issues.
Olorunfemi Tope Roseline
Research Title: Power distribution network reconfiguration to reduce losses and increase voltage stability index.
Qualification Enrolled for: PhD
Brief of Research Work: Reconfiguration of distribution networks comes with lots of advantages such as to minimize power loss, reduces losses cost and to improve voltage profile to mention but few. For my research work I am focusing on optimizing the power distribution network through reconfiguration to reduce loss and improve the voltage profile using linear programming.
Tlhakatswane Jerry Malapane
Qualification enrolled for: M Tech
Topic: Cyber Physical System for Health Care
The aim of an intelligent health care facility is to design and implement smart health systems based on Internet of Things (IoT), Wireless sensors Network (WSN) and Wearable Sensors. Smart sensors are used to sense patient and environmental conditions. Captured information is sent to cloud for real time data processing using different machine learning algorithms. Wearable sensors track vital conditions of patient and are able to send emergency notifications to the doctor and nursing staff.
Mohamed Sameer Hoosain
Topic: Energy efficiency in smart homes based on demand side management
Brief summary: In this dissertation, the researcher investigates and discusses demand side management (DSM), demand response (DR) and artificial intelligence (AI) in the form of Game theoretic algorithm, in order to manage and optimize the daily electricity expenditure and improve the energy efficiency of future smart homes.
Dr Olufunminiyi Abiri
Research Title: Application of Artificial Neural Network in the simulation of manufacturing processes
Qualification Enrolled for: PostDoc
Eustace Manayi Dogo
Research Title: Application of Artificial Intelligence in Water Management
Qualification Enrolled for: PhD
Brief of Research Work: Water is a critical commodity that connects every aspect of the day-to-day running of cities and communities with direct socio-political and economic implications. With projected population explosion, rapid urbanization and climate change induced by the volatility of weather patterns across the globe, calls for efficient ways of utilizing and managing of the dwindling water resource. Internet of Things (IoT) and AI stands out as the core technologies and solution to address water management issues. The present research work centers around Artificial Intelligence (AI), Theoretical and Applied Machine Learning, Internet of Things (IoT) and Data Analytics. The current project explores on the application of AI in water management with focus on optimization, high-dimensional and large-scale anomaly detection using deep learning techniques in the context of smart cities.
Afolabi Oluwatobi Joshua
Research Title: The Use of Fundus and Deep Learning Techniques to Diagnose Glaucoma.
Qualification Enrolled for: PhD
Brief of Research Work: Glaucoma has been attributed to be the leading cause of blindness in the world second only to diabetic retinopathy. A cause of Glaucoma is the enlargement of the optic cup such that it occupies the optic disc area. Hence, the estimation of optic cup to disc ratio is a valuable tool in diagnosing glaucoma. The project proposes a state-of-the-art deep learning model architecture for achieving the cup to disc ratio and also implement an efficient web-based system and hence allow large number of fundus images to be diagnosed remotely in a very short time.
Thembinkosi Nkonyana
Research Title: Big data Quality Issues in Remote Sensing
Qualification Enrolled for: PhD
Brief of Research Work: In the era of the Fourth Industrial Revolution (IR4), enormous research is in demand for state-of-the-art methods and techniques which can store, visualize and analyze the various amounts of data from different interconnected sources. Remote Sensing and Internet of Things are great methods and techniques that provide ways of accessing information about an object without being in contact with it from various sensing instruments. Real world data for these sources is however big and is affected by noise. The present research interest is in Artificial Intelligence, Big Data, Data Analytics, Remote Sensing, Internet of Things, Poor Data Quality, High Dimension Reduction Techniques and Optimization method.
Adeola Azeez Ogunleye
Research Title: Application of Artificial Intelligence to the diagnosis of diseases
Qualification Enrolled for: PhD
Brief of Research Work: An automated Chronic Kidney (CKD) diagnosis system, using machine learning packages and various Python packages such as Pandas (for data pre-processing and manipulations) and Matplotlib (for data visualization) has been designed. The data was cleaned up and pre-processed and algorithms were developed for the predictions of the diseases. Present work includes diagnosis of autism with its symptoms.
Adefemi Kuburat Oyeranti
Research Title: Empirical Comparison of Machine Learning Techniques for Handwritten character recognition
Qualification Enrolled for: Masters
Brief of Research Work: The project involves developing a machine learning architecture for recognizing handwritten digits and letters.
Mdumo Gama
Qualification enrolled for: M Tech
Research Title: Cyber Physical System: Industrial Environment
Brief of Research Work: The research is about cyber physical systems for industrial applications. A Cyber physical system is mechanism that is controlled by computer-based algorithms mostly integrated with the internet and its users. The project involves implementing and evaluating these systems for fixed electrical assets that are very important for supplying electricity.
UJ Faculties
| All Faculties
College of Business and Economics (CBE)
Created from the former Faculties of Management, and Economic and Financial Sciences



Faculty of Engineering & the Built Environment
First in South Africa offering a full range of professional engineering qualifications




