Chap. 2 - Data Understanding

Chap. 2 - Data Understanding

1. Early steps in data mining

   • Data preprocessing
   • Data discretization/representation
   • Data selection

2. Types of data set

   • Record
   • Ordered
   • Graph and network
   • Spatial, image, multimedia

3. Data objects

   • represent an entity
   • described by attribute
   • database row -> data object; columns ->attributes

4. Attribute

   • A data field; representing a characteristic or feature of a data object
   • Attribute types
     • Nominal
       • categories, name or state
     • Binary
       • 2 state (1 or 0)
     • Ordinal
       • Values have a meaningful order/ranking but values are not known (eg: A+,A,B,C)
     • Numeric
       • measurable quantity
     • Ratio
       • inherent zero point - values as being an order of magnitude larger than the unit of measurement (eg: 10K, 5K)

5. Measuring central tendency

   • mean ( Σ x) / n
   • Median = (n+1)/2 th term ; when n is a odd number Median: [(n/2) th term + {(n/2)+1} th term ] / 2 ; when n is an even number
   • Mode - values that occur most frequently

6. Data visualization

   • - gain insight into an information space by mapping data onto graphical primitives
   • - provide qualitative overview of large data set
   • Search for patterns, trends, structure, relationship among data
   • Visualization method
     • Geometric technique
     • icon-based technique
     • hierarchical technique

Imagens e arquivos do mapa