Data Shrinkage for Government Training Course

Why shrink data Relational databases

• Introduction
• Aggregation and disaggregation
• Normalisation and denormalisation
• Null values and zeroes
• Joining data
• Complex joins

• Applications
• Strengths and weaknesses
• Measuring distance
• Hierarchical clustering
• K-means and derivatives
• Applications in Government

• Concepts
• Exploratory factor analysis
• Confirmatory factor analysis
• Principal component analysis
• Correspondence analysis
• Software
• Applications in Government

• Timelines and naming conventions
• Holdout samples
• Weights of evidence
• Information value
• Scorecard building demonstration using a spreadsheet
• Regression in predictive analytics
• Logistic regression in predictive analytics
• Decision Trees in predictive analytics
• Neural networks
• Measuring accuracy
• Applications in Government