Predictive Modelling

Learning Outcomes: 
After learning the course the students should be able to
• Define the predictive models using PMML
• Design and analyze appropriate predictive models
• Apply statistical tools for analysis
Syllabus: 
Unit NoTopics
1

Introduction To Predictive Modeling:

Core ideas in data mining - Supervised and unsupervised learning - Classification vs Prediction -Steps in data mining- SEMMA Approach - Sampling -Pre-processing - Data cleaning - Data Partitioning - Building a model - Statistical models - Statistical models for predictive analytics.

2

Predictive Modeling Basics:

Data splitting – Balancing- Overfitting –Oversampling –Multiple Regression - Artificial neural networks (MLP) - Variable importance- Profit/loss/prior probabilities - Model specification - Model selection - Multivariate Analysis.

3

Predictive Models:

Association Rules-Clustering Models –Decision Trees- Ruleset Models- K-Nearest Neighbors – Naive Bayes - Neural Network Model – Regression Models– Regression Trees – Classification & Regression Trees (CART) – Logistic Regression – Mulitple Linear Regression Scorecards –Support Vector Machines – Time Series Models - Comparison between models - Lift chart - Assessment of a single model.

4

Predictive Modeling Markup Language:

Introduction to PMML – PMML Converter - PMML Structure – Data Manipulation in PMML – PMML Modeling Techniques - Multiple Model Support – Model Verification.

5

Tools and Technologies:

Weka – RapidMiner – IBM SPSS Statistics- IBM SPSS Modeler – SAS Enterprise Miner – Apache Mahout – R Programming Language

6

Case Studies:

Real time case study with modeling and analysis.

Text Books: 
Name : 
Predictive Modeling with SAS Enterprise Miner: Practical Solutions for Business Applications
Author: 
Kattamuri S. Sarma
Publication: 
SAS Publishing, 2007
Edition: 
2nd
Name : 
PMML in Action Unleashing the Power of Open Standards for Data Mining and Predictive Analytics
Author: 
Alex Guazzelli
Wen-Ching Lin
Tridivesh Jena
James Taylor
Publication: 
Space Independent Publishing Platform,2012.
Edition: 
2nd
Reference Books: 
Name: 
Data Mining: Practical Machine Learning Tools and Techniques
Author: 
Ian H. Witten
EibeFrank
Publication: 
Morgan Kaufmann
Edition: 
3rd, 2011
Name: 
Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die
Author: 
Eric Siegel
Publication: 
Wiley, 2013.
Edition: 
1st
Name: 
Predictive Analytics: Microsoft Excel
Author: 
Conrad Carlberg
Publication: 
Que Publishing,2012
Edition: 
1st
Name: 
Designing Great Data Products- Inside the Drivetrain Approach, a Four-Step Process for Building Data Products – Ebook
Author: 
Jeremy Howard
Margit Zwemer
Mike Loukides
Publication: 
O'Reilly Media, March 2012.
Edition: 
1st
Syllabus PDF: 
AttachmentSize
PDF icon Sem 7 BDA-PREDICTIVE MODELLING.pdf187.15 KB
branch: 
BDA
Course: 
2018
Stream: 
B.Tech