MCA-110
Data Warehousing and Data Mining
Turning stored data into findings — warehouse architecture and OLAP, then pre-processing, classification, association rules and clustering.
- Semester
- Semester 2
- Credits
- 4
- Scheme
- 3 lectures + 1 tutorial a week
- Contact hours
- 44 across four units
- Category
- Core Elective - I
What the paper is
Data Warehousing and Data Mining is two subjects joined at a sensible seam. The warehouse half is about arranging data so that questions can be asked of it quickly: star and snowflake schemas, fact and dimension tables, the data cube and the OLAP operations over it. The mining half is about which questions are worth asking.
The order matters more than it looks. Unit II's pre-processing — cleaning, integration, transformation, reduction — is the unglamorous middle that everything after it depends on, and it is where most exam questions about “issues in data mining” actually point.
The algorithms in Units III and IV are a small, closed set: decision trees, naive Bayes, backpropagation, Apriori, FP-Growth, partitioning and hierarchical and density-based clustering. Each is examined as a trace on given data rather than as a description, which makes this a paper you can prepare very precisely.
What to concentrate on
- Apriori on a small transaction database, computing support, confidence and lift, is the single most predictable question in the paper. Practise it until the candidate-generation and pruning steps are automatic, then do the same for FP-Growth.
- Decision tree induction by information gain is the other guaranteed numerical question. Be able to compute entropy and gain for a four-attribute toy dataset without hesitating.
- For the warehouse half, be able to draw and distinguish star, snowflake and fact-constellation schemas, and to say what roll-up, drill-down, slice, dice and pivot do to a cube. That, plus ROLAP versus MOLAP versus HOLAP, is most of Unit I.
Syllabus, unit by unit
Summarised from the approved scheme. Hours are the contact hours the scheme allots, which is a fair guide to where the marks sit.
Unit I · Data warehousing and OLAP
12 hours- Definition of a data warehouse, difference from a database system, the need for warehousing
- Basic building blocks, data warehouse models, metadata, architecture, dimension and fact tables
- Multidimensional data models: star, snowflake and constellation
- The data cube and OLAP operations, OLAP servers — ROLAP, MOLAP and HOLAP — with a case study
Unit II · Introduction to mining and pre-processing
10 hours- Data mining: definition and functionalities, applications, issues, mining primitives, the KDD process, OLAP versus data mining
- Data objects and attributes, inconsistent data, the need for pre-processing
- Pre-processing techniques: cleaning, integration, transformation and reduction
Unit III · Classification and association rules
12 hours- Supervised and unsupervised learning approaches
- Classification and prediction: applications and issues
- Classification methods: decision tree induction, naive Bayesian classification, classification by backpropagation, model evaluation
- Association rule mining: frequent itemsets, Apriori and FP-Growth on transactional databases
- Finding relevant rules using support, confidence and lift ratio
Unit IV · Clustering and outlier analysis
10 hours- Applications of clustering
- Types of clustering method: partitioning, hierarchical and density based
- Evaluation of clusters
- Outlier analysis: types of outlier, challenges of detection, detection methods
How it is examined
- Question 1 is compulsory, covers the whole syllabus, and is made of objective or short-answer questions. It carries 12 marks.
- The rest of the paper is four units matching the four above. Each unit sets two questions and you attempt one, so there are four more answers at 12 marks each.
- Examiners are asked to map every question to a course outcome using Bloom's taxonomy, which is why questions are phrased as “apply”, “analyse” or “evaluate” rather than only “explain”.
The pass mark is 40 out of 100 in every paper.
Learning objectives
- Principles of data warehousing and data mining
- The architecture of a data warehouse
- Various data pre-processing methods
- Data mining techniques
Assumed beforehand
- Information system concepts
- Database Management System
Paired in the scheme with MCA-168 Lab based on Core Elective I.
Books it is taught from
- TB1Jiawei Han, Micheline Kamber and Jian Pei, “Data Mining: Concepts and Techniques”, Elsevier, 3rd edition, 2011
- TB2Paul Raj Poonia, “Fundamentals of Data Warehousing”, John Wiley & Sons, 2nd edition, 2003
Notes for MCA-110
Lecture notes, question papers and assignments filed against MCA-110 are in the library. Everything there has been checked by a moderator before it appears, and an account is free — new ones are approved by hand, which is what keeps the shelf clean.