SEM 6 › Data Mining
ITE3432 · SEM 6

Data Mining

Offline study guides — one per week, built from the lecture notes and lecture videos. Green bar = sections you marked as understood.

Week 01Introduction to Data Mining & Big DataWhy we need data mining, what it is, the 5 V's of big data, and where data mining is used — banking, health and the public sector.
Week 02The Knowledge Discovery (KDD) ProcessThe 7 steps from raw data to knowledge, the architecture of a data-mining system, the kinds of data we can mine, and the five task primitives.
Week 03Data Preprocessing — Cleaning, Integration & TransformationWhy real data is dirty, how to fill missing values and smooth noise, cleaning as a process, integrating sources with correlation analysis, and normalization.
Week 04Data Reduction & DiscretizationMake the dataset smaller without changing what it tells you — cubes, attribute selection, compression, histograms, binning and concept hierarchies.
Week 05Introduction to Data WarehousingWhat a data warehouse is (Inmon's 4 keywords), OLTP vs OLAP, why keep it separate, and the multidimensional data model — cubes, facts, dimensions and the lattice of cuboids.
Week 06Schemas for the Multidimensional ModelStar, snowflake and fact constellation schemas, data warehouse vs data mart, defining cubes with DMQL, and concept hierarchies.
Week 07Design of a Data WarehouseBenefits, the four design views, top-down vs bottom-up, the 4-step design process, three-tier architecture, the metadata repository, and ROLAP / MOLAP / HOLAP servers.
Week 08Data Cube Computation & Data GeneralizationMaterializing cubes, base vs aggregate cells, iceberg cubes, MultiWay array aggregation, BUC, star-nodes and star-trees, and attribute-oriented induction (AOI).
Week 09Frequent Patterns & the Apriori AlgorithmMarket basket analysis, support and confidence, the Apriori property, the join and prune steps, a full worked example, and generating strong association rules.
Week 10FP-Growth & Kinds of Association RulesMining frequent patterns without candidate generation (FP-tree + conditional pattern bases), multilevel and multidimensional rules, and constraint-based mining.
Week 11Classification & Decision Tree InductionClassification vs prediction, the two-step process, preparing data, comparing classifiers, and building decision trees with information gain, gain ratio and the Gini index.
Week 12Bayesian & Rule-Based ClassificationBayes' theorem, the naïve Bayesian classifier (categorical and Gaussian), rule-based classifiers with coverage and accuracy, extracting rules from a decision tree, and Weka's training/testing.
Week 13Cluster Analysis & k-meansWhat clustering is and why it's useful, the eight requirements of clustering in data mining, the five families of clustering methods (partitioning, hierarchical, density-, grid- and model-based), and the k-means algorithm worked step by step.
Week 14k-Medoids & Hierarchical ClusteringThe k-medoids method and the PAM algorithm with its swap cost, the lesson's worked example (cost 23 vs 16), agglomerative (AGNES) and divisive (DIANA) hierarchical clustering, and the four cluster-distance measures behind single, complete and average linkage.
Week 15Applications & Trends in Data MiningHow data mining is applied in four key domains: financial data analysis (warehouses, loan prediction, money laundering), retail (multidimensional analysis, campaign effectiveness), telecommunications (traffic analysis, fraud, sequential patterns) and biological data (integration, sequence alignment with BLAST/FASTA, protein networks).

★Mixed practice exam

20 random questions drawn from all 15 weeks (178 in the bank). Every attempt is a new mix — a good way to test whole-subject knowledge before the exam.