Data Science Course

3.1 Need for data manipulation
3.2 Introduction to the dplyr package
3.3 Selecting one or more columns with select(), filtering records on the basis of a condition with filter(), adding new columns with mutate(), sampling, and counting
3.4 Combining different functions with the pipe operator and implementing SQL-like operations with sqldf

Hands-on Exercise:

1. Implementing dplyr
2. Performing various operations for manipulating data and storing it

4.1 Introduction to visualization
4.2 Different types of graphs, the grammar of graphics, the ggplot2 package, categorical distribution with geom_bar(), numerical distribution with geom_hist(), building frequency polygons with geom_freqpoly(), and making a scatterplot with geom_pont()
4.3 Multivariate analysis with geom_boxplot
4.4 Univariate analysis with a barplot, a histogram and a density plot, and multivariate distribution
4.5 Creating barplots for categorical variables using geom_bar(), and adding themes with the theme() layer
4.6 Visualization with plotly, frequency plots with geom_freqpoly(), multivariate distribution with scatter plots and smooth lines, continuous distribution vs categorical distribution with box-plots, and sub grouping plots
4.7 Working with co-ordinates and themes to make graphs more presentable, understanding plotly and various plots, and visualization with ggvis
4.8 Geographic visualization with ggmap() and building web applications with shinyR

Hands-on Exercise:

1. Creating data visualization to understand the customer churn ratio using ggplot2 charts
2. Using plotly for importing and analyzing data
3. Visualizing tenure, monthly charges, total charges, and other individual columns using a scatter plot

5.1 Why do we need statistics?
5.2 Categories of statistics, statistical terminology, types of data, measures of central tendency, and measures of spread
5.3 Correlation and covariance, standardization and normalization, probability and the types, hypothesis testing, chi-square testing, ANOVA, normal distribution, and binary distribution

Hands-on Exercise:

1. Building a statistical analysis model that uses quantification, representations, and experimental data
2. Reviewing, analyzing, and drawing conclusions from the data

6.1 Introduction to Machine Learning
6.2 Introduction to linear regression, predictive modeling, simple linear regression vs multiple linear regression, concepts, formulas, assumptions, and residuals in Linear Regression, and building a simple linear model
6.3 Predicting results and finding the p-value and an introduction to logistic regression
6.4 Comparing linear regression with logistics regression and bivariate logistic regression with multivariate logistic regression
6.5 Confusion matrix the accuracy of a model, understanding the fit of the model, threshold evaluation with ROCR, and using qqnorm() and qqline()
6.6 Understanding the summary results with null hypothesis, F-statistic, and
building linear models with multiple independent variables

Hands-on Exercise:

1. Modeling the relationship within data using linear predictor functions
2. Implementing linear and logistics regression in R by building a model with ‘tenure’ as the dependent variable

7.1 Introduction to logistic regression
7.2 Logistic regression concepts, linear vs logistic regression, and math behind logistic regression
7.3 Detailed formulas, logit function and odds, bivariate logistic regression, and Poisson regression
7.4 Building a simple binomial model and predicting the result, making a confusion matrix for evaluating the accuracy, true positive rate, false positive rate, and threshold evaluation with ROCR
7.5 Finding out the right threshold by building the ROC plot, cross validation, multivariate logistic regression, and building logistic models with multiple independent variables
7.6 Real-life applications of logistic regression

Hands-on Exercise:

1. Implementing predictive analytics by describing data
2. Explaining the relationship between one dependent binary variable and one or more binary variables
3. Using glm() to build a model, with ‘Churn’ as the dependent variable

8.1 What is classification? Different classification techniques
8.2 Introduction to decision trees
8.3 Algorithm for decision tree induction and building a decision tree in R
8.4 Confusion matrix and regression trees vs classification trees
8.5 Introduction to bagging
8.6 Random forest and implementing it in R
8.7 What is Naive Bayes? Computing probabilities
8.8 Understanding the concepts of Impurity function, Entropy, Gini index, and Information gain for the right split of node
8.9 Overfitting, pruning, pre-pruning, post-pruning, and cost-complexity pruning, pruning a decision tree and predicting values, finding out the right number of trees, and evaluating performance metrics

Hands-on Exercise:

1. Implementing random forest for both regression and classification problems
2. Building a tree, pruning it using ‘churn’ as the dependent variable, and building a random forest with the right number of trees
3. Using ROCR for performance metrics

9.1 What is Clustering? Its use cases
9.2 what is k-means clustering? What is canopy clustering?
9.3 What is hierarchical clustering?
9.4 Introduction to unsupervised learning
9.5 Feature extraction, clustering algorithms, and the k-means clustering algorithm
9.6 Theoretical aspects of k-means, k-means process flow, k-means in R, implementing k-means, and finding out the right number of clusters using a scree plot
9.7 Dendograms, understanding hierarchical clustering, and implementing it in R
9.8 Explanation of Principal Component Analysis (PCA) in detail and implementing PCA in R

Hands-on Exercise:

1. Deploying unsupervised learning with R to achieve clustering and dimensionality reduction
2. K-means clustering for visualizing and interpreting results for the customer churn data

10.1 Introduction to association rule mining and MBA
10.2 Measures of association rule mining: Support, confidence, lift, and apriori algorithm, and implementing them in R
10.3 Introduction to recommendation engines
10.4 User-based collaborative filtering and item-based collaborative filtering, and implementing a recommendation engine in R
10.5 Recommendation engine use cases

Hands-on Exercise:

1. Deploying association analysis as a rule-based Machine Learning method
2. Identifying strong rules discovered in databases with measures based on interesting discoveries

11.1 Introducing Artificial Intelligence and Deep Learning
11.2 What is an artificial neural network? TensorFlow: The computational framework for building AI models
11.3 Fundamentals of building ANN using TensorFlow and working with TensorFlow in R

12.1 What is a time series? The techniques, applications, and components of time series
12.2 Moving average, smoothing techniques, and exponential smoothing
12.3 Univariate time series models and multivariate time series analysis
12.4 ARIMA model
12.5 Time series in R, sentiment analysis in R (Twitter sentiment analysis), and text analysis

Hands-on Exercise:

1. Analyzing time series data
2. Analyzing the sequence of measurements that follow a non-random order to identify the nature of phenomenon and forecast the future values in the series

13.1 Introduction to Support Vector Machine (SVM)
13.2 Data classification using SVM
13.3 SVM algorithms using separable and inseparable cases
13.4 Linear SVM for identifying margin hyperplane

14.1 What is the Bayes theorem?
14.2 What is Naïve Bayes Classifier?
14.3 Classification Workflow
14.4 How Naive Bayes classifier works and classifier building in Scikit-Learn
14.5 Building a probabilistic classification model using Naïve Bayes and the zero probability problem

15.1 Introduction to the concepts of text mining
15.2 Text mining use cases and understanding and manipulating the text with ‘tm’ and ‘stringR’
15.3 Text mining algorithms and the quantification of the text
15.4 TF-IDF and after TF-IDF