TRON07034 2020 Data Analytics and Visualisation

Apply data pre-processing techniques.  

Visualise data graphically, identify feature correlation, remove/retain features based on variable importance.

Identify patterns in the data using exploratory data analysis and clustering techniques

Work on a given use-case to apply the techniques learned

Appreciate the data ethics and constraints that apply to the use of data in real-world scenarios.

The theory of the key topics will be delivered through lectures. 

The code demonstrations of the algorithms will be performed using Python Jupyter notebooks.

Two individual assignments and a final project are given to assess the Learning outcomes.

20% data preprocessing assessment

20% data visualisation and analysis assessment

60% project to solve a use-case provided by the lecturer in the data visualisation area

Repeat the failed elements and the project.

Data preprocessing:

• Feature scaling and standardisation.
• Handling noise and missing data.
• Handling categorical data.

Data visualisation:

• Graphics fundamentals.
• Mapping visualisation techniques to specific datasets.
• 2D and 3D visualisation.
• Python/R libraries will be used.
• Potential data sources are public data sets, Kaggle and real-time/recorded sensor data.

Pattern recognition:

• How to perform exploratory data analysis
• How to identify interesting patterns in data
• Unsupervised clustering techniques such as K-means and Principal Component Analysis (PCA) will be discussed.
• Programming languages such as Python/R (but not restricted to) will be used..