Master of Science in Information Technology (Software Design)

This elective will leave you comfortable understanding, applying and presenting results from data mining.

• Formulating questions
• Features, classification, basic machine learning
  
• Managing data: loading, storing, accessing
• Independent vs dependent variables
• Data manipulation and visualisation
• Overfitting and underfitting
• Interpreting results: What can we say, what can't we say

Algorithms

Most of these will be introduced

• Clustering: k-Means
• Clustering: RBF
• Clustering: hierarchical agglomerative methods
• Naive Bayes
• Decision trees
• Regression
• Correlation
• Neural net basics
• Association rule mining
• SVM and SVM variants
• Dimensional reduction, e.g. PCA

Tuning

How to get the best results, and fix data mining problems

• Clustering: how many clusters?
• Pruning trees
• Outlier detection
• High dimensional data
• t-SNE: what it can and cannot
• Fast k-means

Applications

These are use-cases for demonstrating techniques.

• mining text data
• wage and age, wage and gender
• image analysis
• cultural similarity between countries
• this list is open: send suggestions