Module 1: Introduction to Artificial Intelligence and Machine Learning
- Artificial Intelligence
- What is Machine Learning?
- Machine Learning algorithms
- Supervised Versus Unsupervised Learning
- Machine Learning Algorithms
- Regression
- Classification
- Clustering
- Applications of Machine Learning
- Machine learning examples
- Setting up Anaconda & Python Notebooks.
- Working on notebooks for Data Science
Module 2: Techniques of Machine Learning
- Supervised learning
- Unsupervised learning
Module 3: Mathematics & Statistics Refresher
- Concepts of linear algebra
- Euclidean and Non-Euclidean geometry
- Introduction to Calculus
- Probability, Conditional Probability, Bayes Theorem
- Distributions, CDF, PDF
- Mean, Median, Mode
- Variance & Correlation,
- Standard Deviation, quartiles, percentiles
- Variable Relationships & Estimation
- Hypothesis Testing
Module 4: Accessing/Importing and Exporting Data
- Importing Data from various sources (Csv, txt, excel…etc)
- Viewing Data objects
- Exporting Data to various formats
- Important python modules: numpy, pandas, scipy etc.
Module 5: Introduction to NumPy, Pandas
- Create arrays using NumPy
- Perform various operations on arrays and manipulate them
- Indexing slicing and iterating
- Read & write data from text/CSV files into arrays and vice-versa
- Create Series and Data Frames in Pandas
- Data structures & index operations in pandas
- Importing and exporting data
- Indexing and slicing of data structures in pandas
- Reading and Writing data from Excel/CSV formats into Pandas
Module 6: Data Cleaning- Manipulation
- Basic Functionalities of a data object
- Merging of Data objects
- Concatenation of data objects
- Types of Joins on data objects
- Exploring a Dataset
- Analysing a dataset
- Data Manipulation steps (sorting, filtering, duplicates, merging, appending, derived variables, sampling, Data type, conversions, renaming, formatting etc)
- Data manipulation tools (Operators, Functions, Packages, control structures, Loops, arrays etc)
- Python Built-in Functions (Text, numeric, date, utility functions)
- Normalizing data
- Formatting data
Module 7: Data Analysts-Visualization
- Introduction exploratory data analysis
- Descriptive statistics, Frequency Tables and summarization
- Univariate Analysis (Distribution of data & Graphical Analysis)
- Bivariate Analysis (Distributions & Relationships, Graphical Analysis}
- Creating Graphs-
- Bar plot
- Pie plot
- Count plot
- Line chart
- Histogram
- Boxplot
- Scatter
- Density
- Violine Plot
- Swarm plot
- Distplot
- Pair plot
- Heatmap
- Important Packages for Data Visualisations
- Matplotllb
- Seaborn
- Plotly
- Cufflinks
SUPERVISED LEARNING:
Module 8: Linear Regression
- Introduction- Applications
- Assumptions of Linear Regression
- Building Linear Regression Model
- Understanding standard metrics
- (Variable significance, R-square/Adjusted R-square, Global hypothesis ,etc)
- Assess the overall effectiveness of the model
- Validation of Models
- Interpretation of Results- Business Validation
- Implementation on new data
Module 9: Logistic Regression
- Introduction- Applications
- Linear Regression Vs. Logistic Regression
- Building Logistic Regression Model
- Understanding standard model metrics
- Confusion matrix, accuracy score
- Standard Business Outputs
- Interpretation of Results- Business Validation
- Implementation on new data
Module 10: Time Series Forecasting
- Introduction -Applications
- Time Series Components (Trend, Seasonality, Cyclicity and Level) and Decomposition
- Classification of Techniques
- Basic Techniques- Averages, Smoothening, etc
- Advanced Techniques- AR Models, ARIMA, etc
Module 11: Decision Trees
- Decision Trees - Introduction-Applications
- Construction of Decision Trees through Simplified Examples
- Generalizing Decision Trees; Information Content and Gain Ratio; Dealing with Numerical variables; other
- Measures of Randomness
- Pruning a Decision Tree; Cost as a consideration
- Decision Trees – Validation
- Over fitting- Best Practices to avoid
Module 12: Ensemble Learning
- Concept of Ensembling
- Manual Ensembling Vs. Automated Ensembling
- Methods of Ensembling (Stacking, Mixture of Experts)
- Bagging (Logic, Practical Applications)
- Random forest (Logic, Practical Applications)
- Boosting (Logic, Practical Applications)
- Ada Boost
- Gradient Boosting Machines (GBM)
- XGBoost
Module 13: Naive Bayes
- Concept of Conditional Probability
- Bayes Theorem and Its Applications
- Naive Bayes for classification
- Applications of Naive Bayes in Classifications
Module 14: Model Evaluation, Improvements & Performance Metrics
- Data Split Practices
- Cross Validation
- K-Fold Validation
- Confusion Matrix
- ROC Curves
- Mean Absolute/Square Errors & R-Square
- Ensemble Learning & Model Stacking
Module 15: Kernel Learning
- Support Vector Machines
- Principal Component Analysis
- Ridge Regression
- Spectral Clustering
Module 16: Support Vector Machines
- Motivation for Support Vector Machine &Applications
- Support Vector Regression
- Support vector classifier (Linear &Non-Linear)
- Interpretation of Outputs and Fine tune the models with hyper parameters
- Validating SVM models
Module 17: Unsupervised Learning: Segmentation
- What is segmentation & Role of ML in Segmentation?
- Clustering algorithms
- Concept of Distance and related math background
- K-Means Clustering, Elbow method
- Hierarchical Clustering
Module 18: Natural Language Processing
- What is NLP & How to solve NLP problems
- NLP Feature Engineering & Modelling
- How to process any raw data file
- Tokenizing, remove stopwords, speech tagging
- Stemming, Lemmatizing, CountVectorizer, Wordcloud
- Build models for solving practical read world problems.
Module 19: Deep Learning - Artificial Neural Networks (ANN)
- Motivation for Neural Networks and Its Applications
- Perceptron and Single Layer Neural Network, and Hand Calculations
- Learning In a Multi Layered Neural Net: Back-Propagation and Conjugant Gradient Techniques
- Introducing & Using Tensorflow
- Neural Networks for Regression
- Neural Networks for Classification
- Interpretation of Outputs and Fine tune the models with hyper parameters
- Validating ANN models
Module 20: End to End ML Implementation and Use Case specific discussions
