Explore predictive analytics using R and SQL in this comprehensive conference talk from PASS Data Community Summit. Gain insights into key definitions, concepts, and terminology while discovering successful applications and how predictive analytics integrates into business intelligence environments. Learn about the data science process, methodology approaches, and typical problems encountered. Dive into practical demonstrations covering data profiling, handling missing data, creating partitions, building support vector machines, and testing models. Understand the importance of data preparation, modeling, and various algorithms for classification and prediction. Address common mistakes and challenges in predictive analytics projects, and participate in a Q&A session to deepen your understanding of this powerful analytical approach.
Overview
Syllabus
Introduction
Questions
Sponsors
Past Summit
Agenda
What is Data Science
What do we want to achieve
The objective predict
The data science process
Data Profiling
Our Studio
Methodology Approach
Typical Problems
Demo
Missing or Incomplete Data
Create Data Partition
Build Support Vector Machine
Test Model
Customer Keys
Probability
Save model to disk
Text data
CSV file
Why am I showing this
Plotting the data
Ronald Reagan
Creating Data Sets
Formatting Data Sets
tfidf
Data
Training Data
Results
Recap
Preparing Data
Data Model
Data Science
Algorithms
Classification
Training and Testing
Principles of Predictive Analytics
Wheres the Data
Predict Churn
Problems Mistakes
Questions and Answers
Outro
Taught by
PASS Data Community Summit
