Issues regarding classification and prediction

Data Preparation :

 

Data cleaning

  Preprocess data in order to reduce noise and handle missing values

Relevance analysis (feature selection)

  Remove the irrelevant or redundant attributes

Data transformation

  Generalize and/or normalize data

 

Evaluating Classification Methods :

 

•Predictive accuracy
•Speed and scalability
–time to construct the model
–time to use the model
–efficiency in disk-resident databases
•Robustness
–handling noise and missing values
•Interpretability:
–understanding and insight provided by the model
•Goodness of rules
–decision tree size
–compactness of classification rules

 

 

 

 

Classification and Prediction by V. Vanthana