INT234:PREDICTIVE ANALYTICS
L:2 T:0 P:2 Credits:3
Course Outcomes: Through this course students should be able to
CO1 :: explain the basics of data preprocessing and its implementation by using R programming
Language.
CO2 :: define the basics of classification by using Supervised Learning Algorithms
CO3 :: make use of different Supervised learning techniques to predict numeric values
CO4 :: demonstrate the predictive models by using Neural networks and Support vector
machines
CO5 :: classify the data by implementing unsupervised learning algorithms
CO6 :: illustrate the techniques to evaluate the model performance and various methods to
improve it
Unit I
DATA PREPROCESSING : Managing data with R, Exploring and understanding data, Exploring the
structure of data, Exploring numeric variables, Exploring categorical variables, Exploring relationships
between variables
Unit II
SUPERVISED LEARNING: CLASSIFICATION : Lazy learning:Nearest neighbors, Probabilistic
Learning: Using Naive Bayes, Divide and Conquer: Decision Trees and Rules
Unit III
SURPERVISED LEARNING : NUMERIC PREDICTION : Forecasting Numeric Data, Simple Linear
Regression, Polynomial Regression, Ordinary least squares estimation, Correlations
Unit IV
SUPERVISED LEARNING:DUAL USE : Black Box Methods, Neural Networks, Support Vector
Machines
Unit V
UNSUPERVISED LEARNING: CLUSTERING AND PATTERN DETECTION : K-Means Clustering, K-
means clustering intuition, K-means random initialization trap, K-means selecting number of clusters,
Dataset gathering, Hierarchical Clustering, Association Rules, Finding Patterns, Market Basket Analysis
Using Association Rules
Unit VI
MODEL PERFORMANCE : Evaluation Model Performance, Improving Model Performance, Bagging,
Boosting, Random forests
List of Practicals / Experiments:
Practical 1: Managing Data with R
• Exploring and understanding the data and loading it into different data structures.
Practical 2: Basics Of Data Preprocessing
• Exploring numeric and categorical variables, and finding relationships between different variables.
Practical 3: Implementation of Lazy and Probabilistic learning algorithms.
• Classification based on Nearest Neighbor and Naïve Bayes.
Practical 4: Implementation of Divide and Conquer Algorithms.
• Classification using Decision Tree and Rules.
Practical 5: Implementation of Regression Algorithms.
• Forecasting using Simple Linear Regression, Polynomial Regression and Multiple Linear Regression
Algorithms.
Practical 6: Defining Relationship between Numeric Values.
• Implementation of Ordinary least squares estimation and Correlation algorithms.
Session 2024-25 Page:1/2
Practical 7: Implementation of Dual Supervised Learning Algorithms.
• Black Box Methods, Neural Networks and Support Vector Machines.
Practical 8: Implementation of Clustering Algorithms.
• K-Means Clustering and Hierarchical Clustering.
Practical 9: Implementation of Association Rules.
• Market Basket Analysis Using Association Rules
Practical 10: Model Performance Testing.
• Evaluation Model Performance, Improving Model Performance, Bagging, Boosting, and Random
forests
References:
1. PREDICTIVE ANALYTICS: THE POWER TO PREDICT by ERIC SIEGEL, WILEY
2. PYTHON AND R FOR THE MODERN DATA SCIENTIST: THE BEST OF BOTH WORLDS by RICK
J. SCAVETTA AND BOYAN ANGELOV, SHROFF/O'REILLY
3. EFFICIENT R PROGRAMMING: A PRACTICAL GUIDE TO SMARTER PROGRAMMING by COLIN
GILLESPIE AND ROBIN LOVELACE, SHROFF/O'REILLY
4. APPLIED PREDICTIVE ANALYTICS: PRINCIPLES AND TECHNIQUES FOR THE PROFESSIONAL
DATA ANALYST by DEAN ABBOTT, WILEY, 4th Edition, (2012)
5. R IN A NUTSHELL 2E by JOSEPH ADLER, O'REILLY
6. INTRODUCTION TO MACHINE LEARNING WITH R: RIGOROUS MATHEMATICAL ANALYSIS
by SCOTT BURGER, SHROFF/O'REILLY
Session 2024-25 Page:2/2
