header
SVM-Light 

Tag:
Edit edit   Starstar

Program Information

Name: SVM-Light 
Domain: Machine learning
Functionality: Classifier in supervised learning 
Input: T:The training data ,D:The testing data 
Output: H: A (linear) hyperplane that separates examples from different classes 

Reference

Properties of Machine Learning Applications for Use in Metamorphic Testing

https://pdfs.semanticscholar.org/8b12/c5fd003efd52b798235d97a89a9c91dfd666.pdf?_ga=2.92592733.1914929612.1565157787-500444029.1561960669  

MR Information

MR1------

Description:
Property: $H_{1}=H_{2}$ 
Source input: T,$D_{1}$ 
Source output: $H_{1}$ 
Follow-up input: T,$D_{2}$ 
Follow-up output: $H_{2}$ 
Input relation: $D_{2}$: the values in any column of  $D_{1}$ are all increased by a constant 
Output relation:  $H_{1}=H_{2}$ 
Pattern:

MR2------

Description:
Property: $H_{1}=H_{2}$ 
Source input: T,$D_{1}$ 
Source output: $H_{1}$ 
Follow-up input: T,$D_{2}$ 
Follow-up output: $H_{2}$ 
Input relation: $D_{2}$: the values in any column of  $D_{1}$ are all multiplied by a positive constant 
Output relation: $H_{1}=H_{2}$ 
Pattern:

MR3------

Description:
Property: $H_{1}=H_{2}$ 
Source input: T,$D_{1}$ 
Source output: $H_{1}$ 
Follow-up input: T,$D_{2}$ 
Follow-up output: $H_{2}$ 
Input relation: $D_{2}$: the values in any column of  $D_{1}$ are all multiplied by a negative constant 
Output relation: $H_{1}=H_{2}$ 
Pattern:

MR4------

Description:
Property: $H_{1}=H_{2}$ 
Source input: T,$D_{1}$ 
Source output: $H_{1}$ 
Follow-up input: T,$D_{2}$ 
Follow-up output: $H_{2}$ 
Input relation: $D_{2}$:remove an example from $D_{1}$ 
Output relation: $H_{1}=H_{2}$ 
Pattern:

MR5------

Description:
Property: $H_{1}=H_{2}$ 
Source input: T,$D_{1}$ 
Source output: $H_{1}$ 
Follow-up input: T,$D_{2}$ 
Follow-up output: $H_{2}$ 
Input relation: $D_{2}$:add an example in $D_{1}$ 
Output relation: $H_{1}=H_{2}$ 
Pattern:

MR6------

Description:
Property: $H_{1}=H_{2}$ 
Source input: $T_{1}$,D 
Source output: $H_{1}$ 
Follow-up input: $T_{2}$,D
Follow-up output: $H_{2}$ 
Input relation: $T_{2}$: Permutation of $T_{1}$ 
Output relation: $H_{1}=H_{2}$ 
Pattern:
Insert title here