Page 175 - FINAL_HANDBOOK_20252026

Faculty of Science Handbook, Academic Session 2025/2026

illustrated through analyses of real world Assessment:
data sets covering commonly encountered Continuous Assessment: 40%
data types. Summative Assessment: 60%

Exploratory analyses: dimensional
reduction methods such as principal SIT3020
components analysis and linear PYTHON FOR DATA SCIENCE
discriminant analysis. Feature selection.
Supervised learning: artificial neural Description: Introduction to Python
networks, k-nearest neighbours, logistic programming; Control statement and
regression, naïve-Bayes, classification and program development; Python data
regression trees, or support vector structures, strings and files; Functions; Lists
machine. Ensemble methods: bagging, and Tuples; Dictionaries and sets; Array-
random forest, and boosting. Unsupervised oriented programming with NumPy; Pandas
learning: K-means and hierarchical series and DataFrame; Data wrangling;
clustering. Object-oriented programming; Python
libraries for data analysis such as Jupyter
Assessment: Notebook, SciPy, mglearn and matplotlib.
Continuous Assessment: 50%
Summative Assessment: 50% Data science: Basic descriptive statistics;
Simulation and static/dynamic
SIT3018 visualisation; data mining tools such as
NON-PARAMETRIC STATISTICS principal component analysis and
discriminant analysis.
Introduction to hypothesis testing, sign test
and signed rank test, Mann-Whittney test, Big Data and Cloud case study: Deep
Kruskal-Wallis test, runs test, contingency learning; convolutional and recurrent
tables, median test, goodness of fit test, neural networks; Reinforcement learning;
Spearman's rank test, Kolmogorov Smirnov Network analysis.
test, permutation test, kernel density
estimation, spline regression estimation. Assessment:
Continuous Assessment: 50%
Assessment: Summative Assessment: 50%
Continuous Assessment: 40%
Summative Assessment: 60% SIT3021
INDUSTRIAL TRAINING

SIT3019 Candidates are required to spend a
INTRODUCTION TO BAYESIAN STATISTICS minimum of 16 weeks working with
selected companies in selected areas of
Bayes' Theorem. Bayesian framework and industry.
terminology. Bayesian inference. Prior
formulation. Implementation via posterior Assessment:
sampling. Bayesian decision theory. Continuous Assessment:100%
Hierarchical models. Application to real-
world problems.

176