Description

Course Outline:

1.Introduction Data Science

• CRISP – DM Framework
• Technology stack for Data Science

2.RDBMS (Oracle ) with SQL

• SQL Introduction (DDL, DML)
• Joins
• Views, Triggers and Procedures
• Advanced SQL for Analytics

3.Python programming

• Variables and data types
• Standard I/O
• Operators
• Control flow (if else, for, while, break and continue)
• Data Structures ( Lists, Tuples, Sets, Dictionary and Strings)
• Functions ( recursive, lambda functions, map, filter and reduce)
• Modules and Packages
• Working with Python Libraries ( OS, datetime, system)
• Exception Handling
• Object Oriented Programming ( Classes, Objects, oops )

4.Exploratory Data Analysis

• Basic statistics
• Hypothesis testing
• Data distributions (Central Limit Theorem )
• Introduction to visualization
• Plotting with Matplotlib and seaborn
• Introduction to Tableau for Reporting
• Percentiles and Quartiles
• IQR, box-plot and whiskers
• Bar Charts, Pie Charts, Line and Pair charts
• Uni variate, bi variate and multi variate analysis
• EDA case study

5.Python For Data Science

• Introduction to numpyand operations on numpy
• Getting started with Pandas and operations on pandas
• Sampling techniques
• Data Preprocessing with Pandas (excel, csv and pdf)
• Missing value analysis ( NULL value treatment)
• Data Normalization and standardization
• Outlier analysis and treatment
• Web scrapping using beautifulsoup, word clouds

6.Machine Learning with Python

a) Linear Regression:

• Algebra for regression
• Assumptions  of Linear regression
• Multiple regression
• Feature Selection ( VIF and P-statistic)
• Model building
• Parameter tuning for regression
• Model validation ( Accuracy, Variance, R-squared)
• Bias variance tradeoff
• Case study on regression

b) Logistic Regression:

• Logistic regression intuition
• Sigmoid function, mathematics behind logistic regression
• Feature engineering and collinearity
• Regularization (L1 and L2) and parameter tuning
• Case study on logistic regression

c) Decision Trees:

• Decision trees introduction
• Homogeneity, GINI index and Information gain
• Building decision trees and parameter tuning
• Truncating and Pruning trees
• Random forest (ensembles)
• OOB (out of bag error)
• Cross validation, bagging  and boosting (XG boost, ada boost and GBM)
• Case study on decision tree and random forest

d) K nearest neighbor for classification

e) Model deployment with PMML, H5 and pickle