Course

• LO1 To introduce the concepts of pattern processing
• LO2 To provide insights on different techniques of pattern processing – supervised and unsupervised
• LO3 To provide knowledge on techniques of data visualization techniques

• CO1: Ability to understand the basic concepts of data mining
• CO2: Ability to apply data mining, clustering, classification, and data visualization techniques
• CO3: Ability to analyse data using mining, clustering, and classification techniques
• CO4: Ability to evaluate the effectiveness of various algorithms

Challenges in data mining – Data pre-processing – An overview of data cleaning methods – Data integration – Data reduction and data transformation – Dimensionality reduction – Linear regression – Regularisation.

Introduction to classification and clustering – Decision trees and random forests – Bayesian classifier – Support vector machines – Neural networks – Metrics for evaluating classifier performance – Model selection using statistical tests of significance – Comparing classifiers based on cost-benefit and ROC curves – Techniques to improve classification accuracy Cluster analysis – Distance measures – k-means and k-Medoids – Agglomerative versus divisive hierarchical clustering – Detecting outliers.

Data visualisation – Bar plots – Histogram – Box plots – Violin plots – Pairplots – Distplot – Scatter plots – Pie charts – Bubble plots – Regression plots – Quantile plots – Heatmaps – Plotting covariance matrices – Waffle chart – Word cloud – PCA – LDA – Manifold learning for data visualisation – t-SNE – UMAP.

1. Jiawei Han, Micheline Kamber, Jian Pei, Data Mining: Concepts and Techniques, Third Edition, Morgan Kaufmann Publishers (Elsevier), 2011.
2. K.P Soman, Shyam Diwakar, V. Ajay, Insight into Data Mining: Theory and Practice, PHI Learning Private Ltd., New Delhi, 2006.
3. J Vanderplas, Python Data Science Handbook, Nov. 2016, O’Reilly, Media Inc.