Data Sciences

This page display the Data Sciences skillset that Ryan acquired

The 3 types of analytics

1. Descriptive (What has happened in the past)
2. Predictive (What might happen in the future)
3. Prescriptive (What should we do moving forward)

Overfitting vs. Underfitting

• Overfitting refers to the generalization of the model is unreliable. This is due to the model learning “too much” from the training data set
  • Train with more data
  • Cross-validation
  • Removing features
• Underfitting refers to the case where the model has “ not learned enough” from the training data, resulting in low generalization and unreliable predictions
  • Increase the size or number of parameters in the ML model
  • Increase the complexity or type of the model
  • Increasing the training time until cost function in ML is minimised

Types of Data

• Nominal
  • Discrete categories with no natural order and has no meaning in the data
• Ordinal
  • Has nature ranking or order
• Interval
  • Zero is arbitrary. Time or year etc.
• Ratio
  • Has nature zero. GPA score or weight etc.