Datasets

Record Data Relational Records

Data Matrix Numerical Matrix, Cross Tabs

Transaction Data

Document Data

Graphs Transportation Network, WWW, Molecular Structures

Ordered Data Video Data Temporal Data Sequential Data Genetic Sequence Data

Spatial, Image, Multimedia Spatial Data Image Data Video Data

Characteristics of Structured Data:

• Dimensionality
• Sparsity
• Resolution
• Distribution

Data Objects - represents an entity (instance, tuple) Attributes Database: rows → data object, column → attribute

Attributes (Dimension/Feature/Variable), can be:

• Nominal - categories
  • Binary - 2 categories/states
    • Symmetrical Binary: both outcomes equally important
    • Asymmetrical Binary: Outcomes not equally important
• Ordinal: has meaningfull order but no magnitude between successive values
• Quantitative:
  • Interval-scaled - no true zero-point
  • Ratio-scaled - inherent zero-point

Discrete: finite or countably infinite set of values Continuous - has real numbers, represented as floating point