Module 3: Advanced Data Science

Overview

After completing this module you will be able to use advanced data operation techniques as well as create complex AI models and charts in LOGIBLOX.

1. In section Advanced Data Massaging you will learn advanced data transformation techniques using LOGIBLOX. We will dive into the power of columns as key structural elements of understanding data.

2. Next, in Advanced Predictions you will learn everything about regression in LOGIBLOX, including non-linear regression, and you will create charts based on the outcomes. These are the first steps into Machine Learning. Regression is one of the most popular ML algorithms.

3. Finally, in Advanced AI you will be able to use AI, train AI models and use them for predictions.

Skills you will learn during this tutorial

Flow Builder:

Advanced Work with LOGIBLOX Flow Builder

Advanced Data Massaging - Column Excellence

• Fill Columns
• Manage Time Elements
• Transpose Columns
• Find & Replace in Columns
• Perform Mathematical Operations on Columns
• Concatenate Columns

Advanced Predictions - Regression Excellence

• Building and enhancing Regression Analysis
• Building Advanced & Insightful Charts
• Managing Predictions

Advanced AI

• Build AI Models
• Train AI Models
• Use Powerful Classifications

Missions

Missions are used to help you to practically apply LOGIBLOX tools.

1. Advanced Data Massaging

• Mission 1: Performing Advanced Cleansing
• Mission 2: Performing Advanced Splitting

2. Advanced Predictions

• Mission 3: Creating Heatmap
• Mission 4: Creating Regression
• Mission 5: Modifying The Regression to Non-Linear
• Mission 6: Creating Advanced Chart With Regression

3. Advanced AI

• Mission 7: Feature Selection on the Dataset
• Mission 8: Training Time Series Model
• Mission 9: Using Prediction Model to Create Charts
• Mission 10: Training Classification Model
• Mission 11: Filling Empty Cells in the Dataset Using Classification Model

Data Used

In module 2 we will use the following datasets:

Name used in tutorials (Name of the file) - Transactions.xlsx

Typical questions: What information is provided in the dataset? How are tables and fields related to each other? How should the tables be connected to generate business insights? What columns exist? What is the time reference / order context?

This is a small excerpt of the main data table for the upcoming exercises:

The file business_transaction contains raw sales data transaction records. However, some columns need transformation and completion to build the right baseline for further analytics (starting with formatting and missing information)

Preparation

To prepare for Module 3 create a new project folder named Module3.