Data Understanding and Preparation for Analytics & Data Science

Home > Courses >

Alternative Views:

3-hour Online Course by Deanne Larson
Single-user access license

Our Price: $320.00

Product Code: SC-05-A

Course Exam [Add $100.00]

Description

Course Outline

Exam Details

One challenge in the data science lifecycle is understanding the problem or opportunity, the next challenge is acquiring, understanding, and preparing data for the modeling phase. This step in the data science process is estimated to take up to 50% of the time allotted for a data science project.

This course addresses how to translate the problem statement into data sources, explore the data for relationships and recognize patterns, identify the starting inputs for the model, preparing data, and validating it for the model fitting process.

You will learn:

Review the data science project methodology
Understand data source identification
Evaluate data findings to determine and validate modeling techniques
Review feature selection techniques
Understand data preparation techniques
Planning for data pipelines
Understand data visualization techniques for data understanding and data preparation

This course is geared towards:

Business analysts
Data analysts
Data scientists
Data science project leads
Business subject matter experts that support data science projects

Module 0. About the Course (2 min)

Module 1. Introduction to Crisp DM (21 min)

The Data Science Process
Supervised Learning
Unsupervised Learning
Define the Problem or Opportunity
Define the Data Sources

Module 2. Data Sources Identification (20 min)

Data Understanding
Data Sources and the Problem Statement
Data Source Inventory
Preparing for Exploratory Data Analysis
Data Modeling and Data Science
Data Pipelines and Data Stores
Work with the End in Mind

Module 3. Exploratory Data Analysis (37 min)

Data Understanding
Exploratory Data Analysis
Data Understanding – Data Profiling
Sampling Size
Data Profiling – EDA Methods
Sample Quality
Statistics Basics: Attributes
Summary Statistics
Distribution
Data Relationships
Data Relationships: Correlation Matrix
Data Relationships: Outliers and Anomolies
Results of Data Profiling
Findings – Important Variables
Outcomes and Interpretations
EDA Checklist

Module 4. Data Preparation for Modeling (30 min)

Data Preparation
Feature Selection
Data Quality Report
Feature Scaling and Standardization
Subset Selection
Feature Selection Wrappers
Feature Selection Filters
Feature Selection Embedded Methods
Transform for Data Modeling
Data Ready State
Modeling – Create and Train a Model
Cross Validation

Module 5. Data Pipelines (10 min)

What are Data Pipelines?
Why are Data Pipelines Important?
Data Pipelines and Data Science
Data Pipelines – Repeatable Assets
Data Pipelines – Existing

Module 6. Visualization Techniques (33 min)

Data Description
Data Description Report
Data Evaluation
Data Quality Report
Data Cleansing
Data Quality Scorecards and Dashboards
Feature Ranking

Module 7. Data Quality and Integrity (36 min)

Data Quality and Machine Learning
Accurate Data
Consistent and Complete Data
Algorithms and Data Quality
Algorithm Requirements
Categorical Data and Algorithms
Bins and Ranges
High Cardinality
Reduce Cardinality
Dealing with Outliers
Missing Data
Time of Event

Click –here- to download a more detailed outline of this course.

Share your knowledge of this product with other customers... Be the first to write a review

Browse for more products in the same category as this item:

Courses
Instructors > Deanne Larson
Instructors
Courses > Data Analysis