The objective of this Course is to
- Make a learner ready for solving real life problems using ML
- Understand all concepts via coding for ML implementation
- Participate in Kaggle competitions
- Be ready to learn “Advanced Machine Learning” and “deep learning” in future
- Code in Python and complete understanding of all ML algorithm’s
Module 1 : Terminology & concepts
- Why Machine learning and what is it?
- What is an Error function or loss function like Gradient Descent mean?
- Some simple real life examples on :
- Linear regression – Predicting the Price of a House.
- Logistic regression – Classifying diabetic people from healthy ones.
- Decision Tree – like Google Play store Recommending Apps to us.
- Naive Bayes – like detecting Spam mails on Gmail / yahoo.
- KNN – help Domino’s to open 2 new outlets in your locality.
- 2 Types of Machine Learning : Supervised & Unsupervised
- Machine Learning work flow steps.
Module 2 : Python for Data Science
- NumPy basics for Data Science
- Pandas for Data Analysis
- Matplotlib for Data Visualization
- Scikit-Learn for Data Science
Module 3 : Processing, Wrangling, and Visualizing Data
- Handling Missing Values
- Handling Duplicates
- Encode Categorical
- Normalizing Numeric Values
- Data Summarization
Module 4 : Feature Engineering and Selection
- Feature Engineering Numeric Data
- Feature Engineering Categorical Data
- Feature Engineering Text Data
- Feature Scaling
- Feature Selection
Module 5 : Machine learning algorithms for supervised and unsupervised learning
— Supervised Algorithms – Maths part + Python Coding
- Naive Bayes Classification
- Linear Regression
- Support Vector Machines
- Decision Trees
- Random Forests
- KNN (K-Nearest Neighbors)
— Unsupervised Algorithms –Maths part + Python Coding
- k-Means Clustering
- Principal Component Analysis
Module 6 : Applying knowledge to solve real world problems
Project 1: Consumer complaint classification for a Fin-Tech Company
Dataset source: Kaggle
Project 2: Doing Sentiment Analysis of live Twitter feeds for any current hot topic
Dataset source: extracted live through Twitter API