Module 06-20122 (2018)

The module introduces a range of state-of-the-art techniques in the fields of statistical pattern analysis and data mining. The 'information revolution' has generated large amounts of data, but valuable information is often hidden and hence unusable. Pattern analysis and data mining techniques seek to unveil hidden patterns in the data that can help us to refine web search, construct more robust spam filters, or uncover principal trends in the evolution of a variety of stock indexes.

• introduce some of the fundamental techniques and principles of statistical pattern analysis and data mining
• investigate some common text and high-dimensional data mining algorithms and their applications
• present the fields of data mining and pattern analysis in the larger context of learning systems

1. explain principles and algorithms for dimensionality reduction and clustering of vectorial data
2. explain principles and techniques for mining textual data
3. demonstrate understanding of the principles of efficient web-mining algorithms
4. demonstrate understanding of broader issues of learning and generalisation in pattern analysis and data mining systems

None

2 hrs of lectures per week

Contact Hours: 23

Sessional: 1.5 hr examination (100%).

Supplementary (where allowed): As the sessional assessment

1. Overview. Various forms of pattern analysis and data mining
2. Basics of vector and metric spaces, probability theory and statistics
3. Analysing vectorial data
   • Dimensionality reduction techniques
   • Clustering techniques
   • Classification and regression techniques
4. Analysing structured data
   • Mining textual data
   • Other structured data types
5. Searching the web