ABOUT THE COURSE:Machine learning is being used to solve a wide range of problems across multiple domains. In this era when tons of data is being generated in almost every field, there is a need to automatically analyze that data to improve business policies/future/humanity etc. In this course data preprocessing, regression, classification, dimensionality reduction, and clustering techniques and where to use them will be taught. After doing this course students will be able to clearly identify whether there is a need to use machine learning to solve a problem in hand and if yes, which and how to use those machine learning techniques.INTENDED AUDIENCE: UG, PG, PhD & ProfessionalPREREQUISITES: 10+2+ Graduation with Mathematics and PythonINDUSTRY SUPPORT: All Top industries are working using this technology including Google, Microsoft, IBM and Facebook etc
Overview
Syllabus
Week 1: Introduction to Machine learning (ML), Types of Machine Learning, Types of Data, General Characteristics of dataset, Measures of similarity and DissimalarityWeek 2:Evaluation Measures, ROC curve, Bootstrapping; K-Fold Cross Validation, Week 3:Data Preprocessing: Feature scaling, One Hot encoding,Dimensionality reduction: SVD, PCA, subset selectionWeek 4:Simple Linear Regression, Multiple Linear Regression,Polynomial Linear Regression, Spline, Regularization: Ridge Regression, Lasso Regression, Elastic Net Regression, Implementation of these techniquesWeek 5:Linear classification, Logistic Regression, KNN, Naïve Bayes Classification, implementation of these techniquesWeek 6:Decision Trees: attribute selection measures, stopping criteria, pruning, Occam’s Razor, Pessimistic error, Support Vector Machine (SVM), kernel SVM, implementation of these techniquesWeek 7:Ensemble learning: Random Forest, Bagging, Boosting, implementation of these techniquesWeek 8:Clustering Part-I: K-means & Hierarchical clustering, Dendrogram, implementation of these techniquesWeek 9:Clustering Part-II: Density based clustering, Fuzzy C-means,Gaussian mixture models, implementation of these techniquesWeek 10:Association Rule Mining: Frequent Patterns, Apriori Algorithm, Its implementationWeek 11:Introduction to Deep Learning, Creating Neural Networks in PythonWeek 12:Introduction of TensorFlow and Keras
Taught by
Prof.Manjari Gupta, Prof.Manoj Kumar Mishra
