COURSE OBJECTIVES
Predictive analytics encompasses a variety of statistical techniques from data mining, predictive modelling, and machine learning, that analyse current and historical facts to make predictions about future or otherwise unknown events. Organizations are turning to predictive analytics to help solve difficult problems and uncover new opportunities. Common uses of predictive analytics include Fraud detection, marketing, operations improvement and risk reduction. It finds its applications in all the Engineering disciplines including Manufacturing, Cyber Security, Tele communication and Smart Grid applications. This course will be introducing the real time applications of Predictive analytics in various engineering domains and enable students to apply specific statistical and regression analysis methods applicable to predictive analytics to identify new trends and patterns and uncover relationships.
COURSE DURATION: 4 Weeks
COURSE OUTCOMES
By the end of this course, the learners will be able to
- Apply the concept of linear regression to simple prediction problems, residual analysis, confidence and prediction intervals
- Apply the concept of multiple linear regression to find interpretation of prediction variables, regression coefficients, heteroscedasticity and multicollinearity
- Apply the concept of logistic regression to find simple solutions for classification problems and evaluate the models
- Explain the different methods of improvement in regression models including multinomial regression, regularization, cross validation and feature subset selection
COURSE CONTENTS
Module 1:
Introduction to Predictive analytics. Models and methods, Role of regression in predictive learning; SIMPLE LINEAR REGRESSION: Calculating the coefficients, Coefficient of determination, Significance test, Confidence and Prediction intervals.
Demonstrations in Python and Excel
Module 2:
MULTIPLE LINEAR REGRESSION: Basic concepts, Coefficient of determination, Interpretation of regression coefficients Categorical variables, heteroscedasticity, Multi-co linearity, outliers and influential observations
Demonstrations in Python
Module 3:
LOGISTIC REGRESSION : Logistic function, Estimation of probability using Logistic regression, Evaluating logistic regression models
Demonstrations in Python
Module 4:
PERFORMANCE IMPROVEMENT : Bias and variance, Regularization, Feature subset selection, Feature extraction, Cross validation, Other performance measures
Demonstrations in Python
