“Targeted Support for First-Year Students: an Agent Based Deep Learning Recommender Approach for South African Higher Education

One of the leading predictors indicating whether a student will graduate is determined by the success of their first year. This research problem is a global phenomenon, but it relates to first-year retention and throughput rates amongst first-year students. This places an immense amount of pressure on institutions, faculty and support staff to ensure that the required procedures are in place to assist student transition and timeous student support strategies. In addition, institutions of higher ...

SPATIAL STATISTICS IN ECOLOGY: SAMPLING FOR RABIES VACCINATION COVERAGE IN TANZANIA

A case example of analysis using spatial statistics is geolocated households in rural Tanzania. Efforts are being made to contain rabies in Tanzania, reported in the southern highland regions, since 1954, and endemic in all districts in Tanzania currently. Mass vaccination of at least 70% of an animal population is most effective in reducing transmission of rabies. Current vaccination campaigns in Tanzanian villages have many administrative and logistical challenges. The minimum animal coverage...

Automating Utility Value Intervention evaluation of student writing using Natural Language Processing.

This study aims to evaluate the effectiveness of Utility Value Intervention (UVI) in an African context. To implement an enhancement of automated UVI models through use of an unsupervised machine learning , hence the existing UVI models are still using supervised machine learning methodology.

Artificial Intelligence for Biological Inspired Soft Robotics

This projects aims to present some of the developments of artificial intelligence in smart soft robotics which is natural and biological inspired to construct a soft robotics structure through the adoption of soft smart, flexible, and compliant material. Soft Robotics is now considered one of the most promising frontiers for robotics research and technological innovation. The ability to be soft is more and more a characteristic needed in robotics systems, especially in those that have to interac...

Healing Products of Gaussian Process Experts

Gaussian processes (GPs) are nonparametric Bayesian models that have been applied to regression and classification problems. One of the approaches to alleviate their cubic training cost is the use of local GP experts trained on subsets of the data. In particular, product-of-expert models combine the predictive distributions of local experts through a tractable product operation. While these expert models allow for massively distributed computation, their predictions typically suffer from erratic...