Professional Elective – VI L T P C
Predictive Analysis 3 1 0 4
COURSE OBJECTIVES
To understand the fundamentals of predictive analytics, including types of models,
supervised vs. unsupervised learning, and challenges in deployment.
To learn techniques for exploring, cleaning, transforming, and preparing data for
modeling, including feature engineering, outlier handling, and scaling.
To gain experience with various descriptive and predictive modeling techniques, such as
PCA, clustering, decision trees, and regression.
To understand how to evaluate and improve models using performance metrics,
ensemble methods, and techniques to address the bias-variance tradeoff.
To explore advanced topics such as text mining and sentiment analysis, and learn how to
deploy predictive models in business environments.
COURSE OUTCOMES
After successful completion of the course, the students will be able to:
CO1: To distinguish between predictive analytics and business intelligence, understand the
predictive modeling process, and identify deployment challenges
CO2: To demonstrate proficiency in preparing datasets by cleaning data, handling missing
values, and selecting meaningful features for modeling.
CO3: To apply different modeling techniques, including PCA, clustering (K-Means, Kohonen),
decision trees, and regression, to real-world data.
CO4: To assess model performance with metrics like accuracy and precision, apply ensemble
methods, and improve models through the bias-variance tradeoff.
CO5: To successfully deploy predictive models and apply text mining techniques (TF-IDF, N-
Grams) to analyze unstructured data, using case studies to apply concepts.
SYLLABUS
Unit 1: Introduction to Predictive Analytics
Introduction to analytics, predictive modeling, supervised/unsupervised learning, parametric
models, business intelligence, challenges, and obstacles in predictive analytics, Predictive
Analytics Processing Steps.
Unit 2: Data Understanding and Preparation
Data exploration, variable summaries, visualizations, feature engineering, cleaning missing
values, outliers, transformations, scaling, and preparing data for modeling.
Unit 3: Descriptive and Predictive Modeling Techniques
Principal Component Analysis (PCA), Classification, Association Rule Mining, Clustering,
decision trees, linear regression, logistic regression, SVM, Neural networks, for predictions.
Unit 4: Model Assessment and Improvement
Evaluating models using metrics (accuracy, precision), ensemble methods (bagging, boosting),
random forests, bias-variance tradeoff, improving model performance.
Unit 5: Advanced Topics and Model Deployment
Text mining, TF-IDF, N-Grams, sentiment analysis, model deployment strategies, case studies
survey analysis, fraud detection, help desks.
Text Books
1. Dean Abbott, “Applied Predictive Analytics –Principles and Techniques for the Professional Data
Analyst”, 2014.
2. Eric Siegel, Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die, Wiley.
Reference Books
1. John D. Kelleher, Brian Mac Namee, Aoife D’Arcy, “Fundamentals of Machine learning for
predictive data analytics”, 2015.