Workshop: Exploratory Data Analysis and Data Modeling with R

Exploratory Data Analysis and Data Modeling with R (2 lessons)

Date: 23 Sep and 30 Sep

Time: 14:30 - 17:00

Target Audience: All Students

Medium of Instruction: English

Pre-requisite: Basic R programming skill is required

What you will learn

• The exploratory data analysis techniques to better understand your dataset, such as understanding your variables, cleaning your dataset, analyzing relationships between variables, and maximizing insight into your dataset.
• The machine learning process, such as defining your objective, data preparation, splitting data, choosing and training the proper machine learning model, models evaluation, and how to make use of the best model to achieve your objective.
  
  

Course outline

Lesson 1

1. Objectives of Exploratory Data Analysis
   1. Understanding your variables
      1. Number of rows and columns in the dataset
      2. Name of all of columns in the dataset
      3. Number of unique values for each variable (column)
   2. Cleaning your dataset
      1. Removing duplicated data
      2. Removing redundant variables
      3. Variable selection
      4. Removing outliers
      5. Handling rows with missing values
   3. Analyzing relationships between variables
      1. Correlation Matrix -- the fastest way to develop a general understanding of all variables
      2. Visualization
   4. Maximize insight into a dataset
      1. Statistics summary such as count, mean, standard deviation, min, and max for numeric variables
      2. Visualization

Lesson 2

1. Machine learning process
   1. Understand your objective
   2. Data preparation
      1. Data cleaning
      2. Feature engineering
   3. Split data to training and test sets
      1. Selection of splitting data method
   4. Choose and train the proper machine learning model referring to your objective
      1. Supervised learning models
      2. Unsupervised learning models
      3. Reinforcement learning models
   5. Evaluation the models by model comparison
      1. Selection of proper performance measurement
   6. Make use of the best model to achieve your objective

Practical coursework throughout the workshop