DATA WAREHOUSING & DATA MINING (Professional Elective-I)

Course Code: IT314 B

Credits: 3

Sessional Marks: 40 | End Exam Marks: 60 | End Exam: 3 Hours

Prerequisite(s): DBMS

Course Objectives

• Introduce the basic concepts of Data Warehouse and Data Mining techniques.
• Examine types of data to be mined and apply preprocessing methods.
• Discover patterns, analyze models, and estimate algorithm accuracy.

Course Outcomes

• CO-1: Apply operations in modeling and designing data warehouses.
• CO-2: Implement data mining techniques to obtain knowledge.
• CO-3: Reflect on pros and cons of data mining solutions.
• CO-4: Evaluate models for research and innovation.

Mapping of Course Outcomes with POs and PSOs

Unit I: Data Warehouse (9 Lectures)

Topics: Introduction, differences with operational DBs, characteristics, architecture, ETL, data modeling, schema design (Star, Snowflake, Fact Constellation), measures, fact tables, OLAP cube, OLAP operations, OLAP server architecture (ROLAP, MOLAP, HOLAP).

Learning Outcomes:

• Perform ETL operations.
• Design the data warehouse.
• Apply OLAP operations.

Unit II: Introduction to Data Mining (9 Lectures)

Topics: Functionalities, classification of systems, task primitives, integration with DB/DW, major issues, preprocessing (cleaning, integration, transformation, reduction, discretization, hierarchy generation), applications and trends.

Learning Outcomes:

• Classify data mining systems and identify issues.
• Apply preprocessing steps.
• Characterize real-world applications.

Unit III: Association Rules (9 Lectures)

Topics: Problem definition, frequent itemset generation, Apriori principle, support & confidence, rule generation, Apriori algorithm, partition algorithms, FP-Growth, compact representation (maximal, closed itemsets).

Learning Outcomes:

• Apply Apriori & FP-Growth algorithms.
• Represent frequent itemsets compactly.

Unit IV: Classification (9 Lectures)

Topics: Problem definition, approaches, evaluation, techniques (Decision Trees, Naïve Bayes, Bayesian Networks, KNN), prediction accuracy/error, ensemble methods.

Learning Outcomes:

• Apply classification algorithms.
• Analyze classifiers using metrics.
• Understand ensemble methods.

Unit V: Clustering (9 Lectures)

Topics: Overview, categorization, partitioning (k-means), hierarchical (agglomerative, divisive), algorithms, key issues, strengths/weaknesses, outlier detection.

Learning Outcomes:

• Categorize clustering techniques.
• Apply clustering methods.
• Identify outliers.

Textbooks

• Data Mining: Concepts and Techniques – Jiawei Han, Micheline Kamber, Jian Pei, 3rd Ed., 2012.

References

• Data Mining Techniques – Arun K Pujari, 3rd Ed.
• Data Warehousing Fundamentals – Paulraj Ponnaiah.
• The Data Warehouse Life Cycle Toolkit – Ralph Kimball.
• Data Mining – Vikram Pudi, P Radha Krishna.

Change of Syllabus

Employability Mapping