Data Warehousing and Data MiningR16 semester 3-2 lecture notes

OBJECTIVES:
• College students will likely be enabled to know and implement classical fashions and algorithms in
information warehousing and information mining.
• They’ll learn to analyze the info, establish the issues, and select the related
fashions and algorithms to use.
• They’ll additional have the ability to assess the strengths and weaknesses of varied strategies and
algorithms and to research their conduct.
UNIT –I:
Introduction:Why Knowledge Mining? What Is Knowledge Mining?1.Three What Sorts of Knowledge Can Be
Mined?1.Four What Sorts of Patterns Can Be Mined?Which Applied sciences Are Used?Which Varieties
of Purposes Are Focused?Main Points in Knowledge Mining.Knowledge Objects and Attribute
Sorts,Primary Statistical Descriptions of Knowledge,Knowledge Visualization, Measuring Knowledge Similarity and
Dissimilarity
UNIT –II:
Knowledge Pre-processing: Knowledge Preprocessing: An Overview,Knowledge Cleansing,Knowledge Integration,Knowledge
Discount,Knowledge Transformation and Knowledge Discretization
UNIT –III:
Classification: Primary Ideas, Normal Strategy to fixing a classification downside, Resolution
Tree Induction: Working of Resolution Tree, constructing a choice tree, strategies for expressing an
attribute check situations, measures for choosing the right break up, Algorithm for determination tree
induction.
UNIT –IV:
Classification: Alterative Methods, Bayes’ Theorem, Naïve Bayesian Classification,
Bayesian Perception Networks
UNIT –V
Affiliation Evaluation: Primary Ideas and Algorithms: Drawback Defecation, Frequent Merchandise
Set era, Rule era, compact illustration of frequent merchandise units, FP-Progress
Algorithm. (Tan &Vipin)
UNIT –VI
Cluster Evaluation: Primary Ideas and Algorithms:Overview: What Is Cluster Evaluation?
Completely different Kinds of Clustering, Completely different Kinds of Clusters; Ok-means: The Primary Ok-means
Algorithm, Ok-means Further Points, Bisecting Ok-means, Strengths and Weaknesses;
Agglomerative Hierarchical Clustering: Primary Agglomerative Hierarchical Clustering Algorithm
DBSCAN: Conventional Density Middle-Based mostly Strategy, DBSCAN Algorithm, Strengths and
Weaknesses. (Tan &Vipin)

III 12 months – II Semester
L T P C
Four zero zero 3
OUTCOMES:
• Perceive phases in constructing a Knowledge Warehouse
• Perceive the necessity and significance of preprocessing strategies
• Perceive the necessity and significance of Similarity and dissimilarity strategies
• Analyze and consider efficiency of algorithms for Affiliation Guidelines.
• Analyze Classification and Clustering algorithms
TEXT BOOKS:
1. Introduction to Knowledge Mining: Pang-Ning Tan & Michael Steinbach, Vipin Kumar, Pearson.
2. Knowledge Mining ideas and Methods, 3/e, Jiawei Han, Michel Kamber, Elsevier.
REFERENCE BOOKS:
1. Knowledge Mining Methods and Purposes: An Introduction, Hongbo Du, Cengage
Studying.
2. Knowledge Mining : VikramPudi and P. Radha Krishna, Oxford.
3. Knowledge Mining and Evaluation – Basic Ideas and Algorithms; Mohammed J. Zaki,
Wagner Meira, Jr, Oxford
4. Knowledge Warehousing Knowledge Mining & OLAP, Alex Berson, Stephen Smith, TMH.