More and more products use data to optimize and personalize their performance and offer to the customers. Self-learning algorithms allow quickly address issues that most non-hi-tech companies weren’t aware exist at all.

Seems like every company – big or small, start-up venture or established corporate

  • everyone must stay up-to-date in all related to data-related

This course is an introductory level course to machine learning and data science. It will carefully explain the methodology of analytical thinking. Not only will you know the algorithms, but you will also know how – and when -to start and finish your projects, or which ones are likely to succeed but only with significant extra effort.

This course is intended for IT developers, digital marketers, CTO, business analysts who are taking their first steps with data science, data mining and machine learning in order to provides them the required skills for becoming a productive data scientist in that environment. The course is suitable for people who planning to engage in data science and data analytics projects.

Participant must have:

  • soft skills in programing and
  • Basic knowledge of SQL, Excel and any analytical experience helps
}

40 Hours

Business Intelligence (BI) and Analytics

h

Certificate: No

Price: contact us for more details

Leave your details

Course Outline

Day-by-day Agenda

meeting – 8 hours (5 – theory, 3- practical)

  • What is Big-Data and why is it
  • Big-Data Characteristics & types
  • Challenges and complexity
  • Use cases in today’s world
  • R story, difference from other programs
  • Introduction to R and R-studio
  • Importing data into R, data manipulation
  • Built-in mathematical functions
  • Examples
  • Transformation of variables, subsets of

meeting – 8 hours (5 – theory, 3- practical)

Methodology

  • Defining the business problem or opportunity
  • Defining the business objective
  • Designing requirements
  • Understanding the relationship between causes and consequences
  • Setting up the environment and exploring the data
  • Supervised vs. unsupervised vs. reinforcement
  • Supervised learning: classification vs. regression
  • Overview of model building steps: Data preparation, Model building, Model validation, Model assessment, Model implementation

 

meeting – 8 hours (5 theory, 3 practical)

Supervised learned model R-regression

  • Targeting and scoring models
  • What you need to develop a scoring model
  • Checking model assumptions :
    • Linearity
    • Normality
    • Equal variance
  • Definition of predicted variables
  • Logistic Regression
  • Feature Engineering for Machine Learning
  • Calibration data and statistical model
  • Building a predictive model in R – Concept on the basis of Churn model for cell-phone
  • Overview of model building steps

 

4 meeting – 8 hours (5 theory, 3 practical)

Feature Engineering for Machine Learning

  • Machine learning model requirements
  • Engineering missing values (NA) in numerical variables
  • Engineering missing values (NA) in categorical variables
  • Engineering outliers in numerical variables
  • Feature Scaling
  • Discretisation
  • Assessing Model Accuracy and Bias-Variance Trade-off

 

meeting – 8 hours (5 theory, 3 practical)  

Classification models

  • Explaining the cluster analyses
  • Visualizing the model output
  • Evaluating the models
  • Statistical segmentation
  • Segment Strategies
  • Selecting the “right” number of segments
  • Segmentation variables
  • Archetypical profiles
  • Running a hierarchical segmentation in R  

 

meeting – 8 hours (5 theory, 3 practical) 

Unsupervised Learning

  • Unsupervised Learning and Principal Components Analysis
  • Exploring Principal Components Analysis and Proportion of Variance Explained
  • K-means Clustering
  • Hierarchical Clustering

Examples of case studies